_ - . . . . .r...p. .- .....~Fc.._. . 4 t q. .. . y . . .... - ... r...— E. v . -. .....JJL .... ‘. 4 - ...: . m. 9.291.. . .... .._.. ......l.\.-.-1.. . . . . . o . . . . . . . J . I: 4!. . d P. .... s ......"15 1825‘..',! 3...»... ...: I..-”v.3“. . . ...... .... .u.... . “.... .7 z, .. kw...uwu.w...u.m”a.uh:.....fé. ...1 ...D’.......:..u.....o...:...2..fiu ...-.12).}. . . 11.72.-.. . ....I... . . . . . . .. .. 7.1.334 . .4. y .. .V. 9 Q a .. t .1 D “$84.33.”. Li....ownfl‘..a . {It}... .13... U: . . ......8. . _ ....7. ..u. .fl .... . . . . .... A. v... .... -.. .o :19. . ..., . ... .9334. it. 9...; . r339 .... ... .. . 09... . . . .1...th 0339‘} ... I... S.”....~...O..%£.IA. . . . . ..7... . . . ......39? 4.. ...... ... Z... is... . . .. . ....8. . . .... O. . .. t . .. . . . . .. . . . . . L . . :51... 1.25“”. .. ..‘r 0.... ..9 2 . . .... ... ... .73.. ... . . .....oni .... 2. . .. ...-t... .... . .. . . .. . . . . 1b....2f'l... 3.1.1... 3.53).. afiaqh. ......olo...§.. .. V......3’1 . . .0... .tl... 3: o vtio... . . . . _ . . . . .. . . . . . . . . . .. . . . . ...... .I . .324 a! .... 1‘ 3.... a... .fik..3\.d .... . ....Q: n. v.1? o..?c.!-.’. .. . . . . . . ,. . . ._ . .. . . .. .. . .. . . . . . . .. .o...’..’& ..hknul 1“. ...“.F c l .J ”Gilt...” ...!..-.’...|.)’..J......0nvn....nuon,. . I.....I.. i.....rvuluuoo..1.. . . ...—... ”nun“... . . . .. . . . . . . V. . . . .. .. . ._ . , .0 .9... ....Z..o\. ~ diva. . . .. . . . .rMbDWO... unlit“... Us‘. .- w.. '1! m . ...Lcuo. _ .. . o. v.4».v...‘d:.oo l...‘..~.o.. .00.. .. .I.90¢ f... .31.. .l“. . P .1. o. . . . . . .. . . . ..O‘ n. . . .1. O O ......5 08. ... .- -t\...f..n§|uo A”... us-"qfin. . q ' O... 0.. I . O \ . . . .L.‘ . n .0 . U. U; .0- 3’1. 5 . ‘ ’ I. I v.< 1 I ‘1 . ...:I o. ...-0.. . .. ’ ... .. .. .3... . . ...KL: . . It... 3 .9? ... ..4. 0- .... ... . 7.. g a. . ‘90.“ way A. no, . “arr“...lws.’ 3 b. .1 . J 3......- . . . . 9...... V _. . O .. ...!) .I.L.z..I.Yo. (Q...QJ."... . I03... .311! ..‘t‘....’6‘..cfl.o.a.’.fi.).. 'l’..\».l.. 9% 4'. ‘ .... ..4. . . . .. . .. . . u . . ... .. .... \‘x .. u . ... . . . . ... . . a. . .... ~ ..cf'. 0- o (a ....'t I....w .... . on»... ... ....3. IV. . , . . CO0.“ ”VQ’OJIXJ. -. g .' .....HQ’..3‘33‘1.”.'.I. 0. Ii“. ”..."...9' 0-74 a (... c. ......4 ... . .r'afll O.‘ IOU'IL. .0 4’“ - 0““,n”, .P’fiw‘gh‘.‘ .‘ooh. I ol‘t‘ Lunnbkrfibt’: b ..I . it... 50 . o .43... O. .o ... ......u. .3 ......Q.tv 3.... . . . . ... . ......OP ' ......7‘.....’.:’ .3? I; c.3001 1,2. . . . no... . ‘3‘... ‘30.»..‘IQQ .4 p. ...... 5V. . (gov . 0...... ‘) su‘dfil‘ll 00.. .....‘H . a . ‘01.. .... ...... ...o. ... . .Vus. - ...“... .... 9.”... , . . .... I. .v.... ..4. ...-3.. . .. ...I3'hr .vgvo C . . . . . 7.14'1. ..v”.“.&.‘!. ...;a .. a vrln’. ‘15., . . . . ...... . . . ......i. . ... : .... .9 7......3. .1 ......l. . ...mf ......) .. 3. 3’2... . .. .. .... . a. .9900 32!... ... .. . :}§;.S.. 5.... .D 18...... . . . 23:? ... ..3 (11. 2:19: ... .....-lv.|-v. . $3350.00. 10...... . . .. . . . I..“...Mr'....".4’ . .V . . . I....L. n‘.o.de‘.qka..fd'-§ b... . ....Cvthlzo: : ... ..w ... . . .. . .1th .....Q: ..l. : .,.........r)..c .... I$ll. _ . . . . . . . , ; . .....0. .97. . ... 90921.... . . .. _ . ii tl’.l....)’oil.lmfi 3 OIL ...... .... 98...; .. . .Q 4 .r! 3 '.v!w\...:2' .. . o. .o li...co......o . .....- . . .. . . . .. . Ur..." Mu. . . . . . . . 3.1.7... ,‘2 .4. I..o..!.. I I... .1. v . .sQ...‘.c 2. ..v,o‘l 0 . ......" .. . .5 VAV'O. . ..'~.Oudo'¢o..o‘.1..o.2'...n 310.... . . . . . ..“D' I. . .. . . ... \a L..). ‘I ...” .3 h . . . . . ........\.o. .... ... I..... in..- ...3: ....- L. . . an. . «r 4.. .. )1 . . . ......- 1 .s! . 30.? II . .... ... .0... .... ... .. . . ’..O.,.4. ...-.8 .9. I o .0 ...-......o? .. .... . ; ...... v. ...-93 . a .0... .. .1. .1 . .... .97.. ... .I..!. v.01; .0. V’ .03.... . n .c -..‘I ... . .. .. . . . A 13.1. a: a '33,... D .- o 3 .voo. ".'.w.nr.“uw”. . . . .vr....t....m. .. .3... . I.. $.35... .. .3. . u..1......o .....o. 7... I... .o ........t..o.. ... 2’1....118I.21..\’.. 13.0.... .1! ...”...Y. . “Fifi. 2.3.... .l“!.9.?..uva.!.9fl...la.uu...“$03!... 8 .. . . . . .. . . . ..L. ’3‘ Saw o.v.l%.£#. Ifi.%a.z . . . v I . . . . . .... o 13...... .v o ... ..o .-..o .. ...lo . . . . . .— rlo. ... Cs. 0 . 0n .. .... . . Q. .. . . . . . . . . . . ...v. usilwflvflmr 1.. I m... ...! .... .v “......3‘01fiuofioad... 0. ”Huhbd ... .IVQ....3.... .....r. J..bh.02l...'.h. .010. .... .0. ......OO......IO.§. .33. .v .. .. .4 . ......10212 ...... .I I a. ..29.....4ov...-..... .... ‘1‘. . .. . {.97 $5.13!... ......LLaustl. .0 ...”... v n . . .. . ...: ,5 .. 09:3..‘9‘V‘03.r1 ...-J';\.Z. .... . 0.... . ... I. V . ...-. 39.0.21: o‘ ......9..'o~.1o.\0. . 03.... ‘3... 3:7 .J 4.. ...? . ., 5‘17 o 0.... . ) . . . , . . . I."~ ..«r; l¢.~.¥4. ‘1‘! ‘92, ‘ flat .4.. v; . . . . ..., ,.....:.....!.\.h.. 5...... -.. ... r. ...... 3.. 2L“... . 12...»? .....v... ...... ..>l._..tl. . ..-... 9.21:...‘095 ' .l:lo......0¢ 1...... .1... .... ....v..-‘ . . .. ..........h..\ ,0 . . . . , . .. . . . . . v ...!!233 ...... ...r . 51.. ... Q ....v. o 3. . I..!E... 9,16. 0}. I. .n. Ill .0“. ’5‘ v V. 3..".3{.9....|.\. , . 1...... .... ....391; c... 4."...OAOQ . .. Q. .0....-6... 1.1.... . . . . . . , so. .I”..*. Q.;\. in ..I .01 ...: . . 0.9.1.370... ti... . 4...... . ...... ... ....T: ... -4.. 7:... -...‘O . . 5 .... _. ’ .... Q. ...... 101%-.. V ... . .C...)!1l. V. L. O! . '0 o. I. 4‘ ..,..g . . .3 ....‘3.." . ..O xvi... ...... .... . . .38-... ... 3 .r. . ‘1... .05.. . ... . ...... ...-3.6:... v6..K...........?J ...: ...-.13....) ...... . . . 3 . o. . . A... . . . . .. R 2).... .§ x J .. .... .0 ~ . .4 0,. . 2...; o a. , . ... lm. ...... . s . ... . ... .l'...... .1. 3314... 7:17 ...! .91.. ..Ja.....’o.’o.:0... ...9.‘ 2...... .. . .. .0 vint’vngf‘zo- . .4”... ‘A‘. 33-. .9. . ... . . . . . , . . . . . “‘33-... 5 . , '5‘? ...)... .3... .0 a. ...... . . . (9. vi. . .....‘.r~ I. .. c ... ....97‘5. ’1»..I.o .33.. ...-...... 1. V. .. c 09..., 9. .1..n..1.9o.).«v.... “VJuJN'V'VJoi . . . . . 9...)!bl . .‘-(v Q” _o O ... . t. 59.... n n. a..- ..l".“.otJ..o. ... “...-9‘. ... g or}, ... 9..., 2‘1... .0. 5.9.9:... .37 ‘- . one": ... .1«.sm§c.'§:u.. ),un Q. . 00. Q O. ..".I \‘v’. ...‘t‘rug o....”...f‘.c.o 9.3"}: .‘2..O.s.....o3".‘§“’.. .?1..~..'3J’ .l ’O.‘ . I. .....‘a...¢.~| I'rf’ Y .'l| z 1., O 5.0 . 0.... 0 c a. :- ..-.r1t C .031 ‘ t .. . I . If. OH»..MI'H...IOV“"QK.! ..vlo. ..n‘lso'o‘,’ g 9 . v 1 . 4. I - . 0t u \ .0 .. . o . a . ~ \ h . . . ...n.l..o.l.....,.- .lnwn-‘aou.voo. .o .I.u .. . ... ... v...6\v|¢.... l§.~..o~..".ll .0 .s} a: \’}louuov.ow..‘..‘0’l“. VAN-DD."13I.. oflzsgfi§0.t to. . \ iv .‘\ _. ...l..... 5...... v 3.91.; ......HWm. . ..9 $1.. .... . ..vv... ...... ...-...: 1.... ... . . . 2.9.. . 035:... ...... .1....&r;‘b.o.. ...-.3... .-.. .3... .2 . 1.2.. '3 .r: .. a l. . .. .1. . ....» 313.108....“ Q... .3 ..‘l O... . c 3 1.0).. cl .... .. .4.. .0 ...... i , . . , . 4 . q Instr... .01....)3..I..3va.. .30351i...‘ C P 3!... 9...! I...- o s. 1"... Ottawa“... . . .... ..l . .I ...... . ..olvo. . 11... ... 7....921 .0... III! I... ...-... 3373:!!ch X ...! . . . . . . .. . . . f 1 . I l...|..... ..Vmiz 09.!" ..‘znvtfikrloh. ...]..J‘Q . l. ...... . . .... ......t... o . S. 239..., ...2\:.....1........1... .2..0!9'r.?.. ....-tl . . .l: .. 4 ..fl . ,. . . . .0... .35), 3.9%.. . 0‘3 . a 9.. . ...... .3. §St3ifgo -9. ......Z. a. . O. . . .. ......o . , ... ... .9 .. . .. ...Q._1}>4..l17... .91?.4.v!!9 0.5.9.1... .. I... [icy-...... . .. 90.1.9. I ....9.‘th 9?... ...... ...»! .... {20.3.1532139 , 3.393.319 . ...-l ... $23....Ih31. o. .123...I.....c...‘...:’..9.......l1i 0.... 61......5! .4.... .. u...............\.. ..... . . : . i 5.....- - ...-BA... 02:... ‘ .I....It. Al‘.)...u s ...r. r." . k .9 Inc. $1.. .n 3. ‘1‘... . . 0.3... v 3.9.3:... u . ...... ... {A}. g.‘.fi:.‘. ... o. I.. .‘lt... . ... ...........1Ji..t......o_l ...... 19... :9... I..... . .. . 9“... ......o . . . . . . ...)I.l\-2 .. ...; «z I... 3.3)....1‘... ... x. . I" .4 . .590.-. 0.... . 4 .... O: 9.. . . to... d... . .. I... 4.... .... 0). 3.2.3291}. .. .36.... .v.... . .1... . . . . . , . . ... .0170.- ..xb..‘ .s. .0 161?)...‘t: I). ‘-\:F.F .9. or.‘L,slol\. ._ 3. sn- 9. ...-l. I. .Q I . , . . f.....;..'t..........v. a .... .o .. ...I.s.'...m . 3.... ...... .1 .....m. L. . .13.. . . . c. . ...?Ot £03 E.‘I.prt.’ . 0.99.1. ...-8“.- .O.C..q.....nn..o. .... t I .... 9.0. . . .. J ......o 3...... ... ..1. Ci: . .31.. ......c.....!t) o ....‘uor7. 91...... ....4’ll..lr..a D . . . . . . .. . . a... . D ".14).; ..K .\ '.o’..,.8~1§ .39... b I "I II ...-13¢..u‘. ”I“. 1 Li." 3.. .4)... 0|... .\.. ....o . . ......w I..)! . o. (...... ......140. .. ...: .. ..l....:.... .....K. ...! . . . . . z . 1...... . .ol. ... ”......1. ”3...? ...!.......|y.. 1...}? ...... .19.. .....Y. O . . .u‘... i. ..C. 13......201 3-. A . 3% ~ . \O‘J. ..- ..) 3". ... 2 . ca .. . :fi ...... .... v.- llI-.'i-‘v .fiian. .. .0... :9... .. .... .... .. . . . ...u. ...: .... ......urars «...... ”...... 2...... .9 . ...... ...... ...... 3229......” ..r:.....:: 1...)... . . .....3. .o 2...... O . u 9 .... .... v.09 .... ...... 113...}... I... ’31. .... ......v .. ...... 11.3.9... «.... .... . 2...... rot .IIJWszt1thI3 ...! . . .I .... ... 't“!h..9l..ufl".t.‘..e v. . , . .Y. o ...3 ...Ov‘...8~ .. ...... .t....!.!’).. v 3...). 3..-... 19.. . .. o . .. 1.! ....It‘..\\0 .... t u’. ... 3.3.90? .1 , 1. 2%....0 ..C...«...I\. . 3.. ...l l,.. ...}... 9.32331: 3. ...-0. .3.v.... .....o. . .70 .. . I.. ‘2! . . O) ...“ t. h.- .6? 1 3‘1)? 2... To“ 2 9:50.! 33.02.... 7.. in .. . 2-.....623’2 , .207!!! ..o 23...... 3\9~|’.... ....!.§.n.vi302 . ... 1...". 9.... III 1.... .... . . z . .o . , 35.”... O... 70...... .‘I ......)lv ‘0) . o x. (Q... .2030... '4 ii... . . .0." . . ....Lél. 3...... I... O. .Ikl...‘ I. 5.9.9.1.! .0. ..JC‘A..YI!(S..3!I.. .ll. .. . 3;... 9:1 . . ...!!! .. A. J... . .....Q...9 ...... ...-....«xi . ......vucvflrfllpwf? It». Oihr93h.~qao1t..3. “Oglk .550... )SI‘L..(O ’95.; a... ..-..o. 1.... . . 1:3}, r...... .- . l . . .v.u.....s ...‘o...3...3\.,l\ [2.1. I..... .15) ......3 .23... 7... .39.... 2 V ...P!!....! ... . . . 9'1}. 10v... .. .1. ... on. ’9. . . . .... s I... ...... .1 . at“?! o9..l.. ’9‘, \ s... - .3oo"61.u.l5’.a.a 3.. 02".. . 3.” 0‘. I....J . ...: .llo...\...J..d~ . .1 . 9...“. n... x.vm9'fit3.v.o.:....ts.07It... 31‘-..‘ CI .....Ilo...u...c. 9..!..3Q.E.ris -.. ..i ...... .... .9... ....ul.!..‘.s.!~.“ ... ...? ......Ic. .9 ll.....h.a...9 «It! ...-... .9! "- ol .5 .... .3}... .. ... ...... ii. 93‘}... ‘1)? ..lys‘uf... Q ..SIA. ....I.....-.r.. 1» 3.. . 3. . ....S 6. -. .87).!18 ,. ...).-.fi ... .... I... . . O. ..O. n. r... . ... .32.... 49...... 1.2-; ..l..\ 13...... . 213.955,. bu 98;... ..?t niJEOIQ...‘ .02....)2 Q... .. ... .tfi~.\l...... ...-.53....‘rd. to. 11’s ‘13.!!!‘5‘! I ..I . A... .....r O 833'. .2120... (.41, 3k ... .... .0303..3..9! 02.....1 1...: 371...... .28 1.8... 2Q... ..9. 1. ...... .. .323... ....‘93..!.-v0 ...| .. .. ...». . , . . . . . .. 1"ouivasw 2.3.1.... 'u.‘9>...‘.\9}. \- Dc‘. 5? J \33“ a. . . .a I o 4.335. .. w”. .. ...... .193..81X..t.¢'t:). 2 ....TSISEI’.‘\....::£I... Io........ o I..! .... ...: 81......9n. ..........I ..i...t.o:.~3! . . . ‘9 ....)Ii‘OVV . . . .ritl. .....I..V£O.u.l..nl.v 1))... I..... u.‘."..‘lo. ...." I 15m: ..Q. .- o . .. ... oh... O0. ...: ...-.3 9.0“... Q.- .-‘4-|v7....u U. .‘c... o. 2": .... . . .23. . . .. . . . . . ...... 9’ .- o‘tI’O-.;. 3'...‘.‘z3.\". ta? , ‘u‘h . '5’ .§’-‘.-~ .. s. I .u ..a ... 3”». . 10...... 40.. ... . .4)???".tIY’..§38!...k.q ...... . ifiz..\9 2.5... 1:90., . .. .. ”’9. ll. . .9‘ I‘P‘A. 1 3.2V .. ‘21.! . ....il'fifiv)... .. . . I I . _. o . it . 2.1.. . £53-91. ...... . 4 . . Dali-03"...Iztfll 4.. .;>'!'$ ..4.... ..- . .00. . . ._ 3 a! .. .§.?... .0 Di rinsilt .8 $9.. .. . 0-...Dun..do ... oob.’~u.l. .7101. '0. u .. . .- .Itl; :- r 5.! .. 2:311. . \-‘Q..v§...vao.."..."§au.9|b. I. u . . do... 0‘. ......‘n..l=.i‘oc§ .9... "I 1.3.5" “ i. o .3010... .....n v d. ...l, .. data-...} . ' . . . (zxuwcl... ”...”..IV . , ....o...to.... \.. 6‘ ..vt...’ .3. P h. . v .9003: .....I... IMH’“.OXU .10.! ...... . . . \"o.7’ vutuurdvoo’n.p'?r.iofl“ '1‘ 1i.;.._..‘ .I .. ...-‘5‘ ththkolhfirQ Id . . . . . . . . . . . . . . .... z t . .1330... .. - . . .. 7... . . . I..! _ \oo. . ... . .. ‘0‘ . . . 5.. I..... 40.3.. . O .‘ no. 0.0. 3... .~ I! .1... Ito.- .‘J. 0.. .. 0’31? 6 o .\_.o- .0... ... co .... :33‘ . . . . . .~ ‘37.. ‘Iwoi’ounoJ: .... . so}??? ...I v .0 ..ld. 110%. 4 s. .t!"1 .69.. O. Q‘X‘...‘c¢§lfl.’|flfln’xv‘ 9 link“ ”gun's“, othhfl‘)-..~.. Rhyr. .....9379: .... 0.! ...v‘.d....) II o. .4 -... 2...... ... .9. ...-...l... .... . . . . ... ...»? pl"... ......1'........:.!.‘.1...X .91....331II‘... . :1“. £1.11 . 14.53....322323. 3‘9 ...-‘0! . ... .....o .. ... . ... ...... 2.331.... b. .. . ..9. 3933!... x . . .. .. 39.321?! t I . .... .- ...! .3... .blé!!wv 2’) in... '5 J29} S...’9...”.D.. . J. L»: OI. .. ...... ...A.. {3167:1- . .... i.- ‘I... In} .. ..-. . . . ..e. 1.3.2.0 .f.‘.vti9!...... . ..Qo .. ......rrv. .... L.).!. 0.. .r...‘ 159... .03 19.2.... ... . ......11) .. at! A . .... ... .’ .9....I.-. ......I‘i’l .. I..... ‘. d).....h ; . . . . 0%... ...q. . a’.‘...:... ...? ...-.21.. I 3......23 ... vii-.1 (- . 4 ‘. I‘ Mr...“ to. u... 9 ..O a)! .3}\IC‘.. o~.“.$1!'0.|.1¥11£€. 3’. 13...! 3.3.1.0... I... 11.....391... ... . 3...... 1.3.3.... I11 ... ...... . C9. . .. 91.... a! . . 9... .. .. ......g- ...... ...... 3.32.1-9... .... :.v........19¢.3.‘.1..9-. .n cc... 0... .. I!“ ‘0 . 0...? .I ”no! .t 4.. :41..- .‘ol..."q.!..§s (Sig: .\ ...... .... . . ..‘S .Q! .721! .O?‘ .....-O..--. ’8‘...‘ .. . b. . . )1. .-. § . 9 9:55.... t‘i!!!q.7o..oo¢'b.... . . ... air}. . . .5... .11... ...-315.01.32. cu. .... .9 ..3 W ...-o ...,LOI...‘ ‘v”\ll .i....'\ '.V 3.1.“..1 9. . ...bul: ... ...r........0...l..>. . . . 1233...... _Q...n....o 1....t 1.730.029 .... . ....VII ...-9... .. ‘t’..t..’..t$o:vv -..-..ut... h. v .0 cph.._v.l...hO!-l! $3.01,... - . ...: a... 21.39.... 3 zeroit: ...-.... ...... ... ......to. I... ....v . .... 3.. ..fl...io..£ .. .. O. . O... ..v.‘é.v.u.a :3 . P... ...“...9.“......2....IO...91.,0..9 .... $139": .2)...u...1.u.......t'\..!tt 2...“. .LPWJM. ...hh... . . ...sfhfsl .0 . ......flarxll ngiifl .0.“ . 33“... 3...... . . .oll) ' .. ,- ...l a . (\QA v... I co. . d o. n . . .. . .v . . . . . ,. i ..A. . x r; . .0.-. 9.... 1... ......2. {.36. ...? .8: I. .3. 9 0.... .301. ..flold§.8..l.n“3.u..ra..nvu. .. ” .... I. ... 0.. . . .092:7I......0 .0..I.0 .. I.."— ..i... ....K...l’.’ll (1.. 29.593! 0.30! ...I O J‘Q. diurfivsuflwo. 4 I J’zulfi’v‘I uizfitflfi.tinl.. 1“.“ .‘ ‘D‘ho‘ -iiw. ‘13. ...... n... .36 ...... .. .x .I. . £ . ...!9.... .....tr; 333...... x..I:IA.)..-.3 9... . .. . . . . . . . ..zb.‘......-0..lb" IQ... .... .93... II bflib!‘ ~13”. .3). .. 3V; . I 4 . . ..‘a .... .. ,2... ...-2:20.13 ....~ 0. .2. ..Q‘l..5...l ... . . . . . . , . . . . z .I ....9]!!!Qb.'s...'.fv:. ...} 0.1”‘. . -I .\nt.o,l 1.. .1... 9. . . .. 3-1.3... . ....vv... '3‘ 0-. .6... 3“? O . I _. .9.\..|.u.o .1131) . It‘.’ ..- b... . . . . . . . . .. no 01‘. ...-".1 ‘56 1.. V}"?.-. . .;;.5.“ 9.0.!“ to; ...-3 ‘ti. .....o.-9.....I..333 .... 3.35.1013... . ....nyo.‘ . ..‘..a.a...v.o... ... ...-U. I. .. ..Itv that Co. 930.“... ‘....o...... . v . . . .0 . ‘99.:‘N...’....!» DDI.I.¢W§.2$Q§.IR \r' b... .00.. 15:52.! I 7...... O.- 0........vll .J .5... O .0. .lotoo‘... 6!. P-.. . so‘.1.o. 9s C.l~vl3 5 #11... s. 'I’v4..!).. .. . . . . . . . . . co... ...-..‘ I. . ) $9,3".¥fi I! .04.)... .5.§’<.’o.=..ozg ...)qt ’9‘: .. .id‘il .3 ...-l... . 6.91-...01...o~7 $1.... ...J . ..Ig...oc . a: ...: A4. A: . 3...... . .19.... . tbs...‘ .D.» ... 5.....a: . .. . .. . . . . . .. ..3’. 8:9 0...“. ....‘.‘.\..9 D . st... $36!". m h b. .16". .9. ....ii. ...... .... ...... ...... . .. .0..£:.¢.!.... 54......3f. ...... .. ... . .T.. ... 31......9! .. .1382... . ....u ......c: . . _ . v . : . . . o . b ....s. Cufl.~.\969..cl ..a ‘9‘ ll . .. .. tr. O .l. lli)...’ . .... .10..¢..11...... ...! 0‘63... r. ... ....9 .23....636 909...: ’2... 5!. £123.... «”199?! u. lo'. .....1‘ I. 03......- 03‘5..-‘ . 9“..\ . .. . qr. . iv .....v. .... n ......l. ...l... 74...). 30“.... ...... ....771 8.1.1.... ‘ ...-0...}: .....v... a... I ... . 0 p . o c..x.¢1...‘x I..-v) 1.. Fight-flifiéadi Si)??? ...-N .915 ,I .7211... ... 3.4.3.. ...-...... .. 111.2,... .l‘v‘lxu.l. 13.3.13. . .31 .‘ $0.). .14.: :I D“ ’a‘. ...:Qlunt. Qifi‘...:r nggfi: 53“} 4 .I '3» . 4 .0 ...)... it. ...; (.....A..‘. ... ... t I . .... I..... . .I... ...... ’72.. 10.0. .r .. 30,-0.2..2 ‘21:... ...!!561 1:1...| 13‘. . ...! 1'7... ......‘1. Praia... .. ..v?!.a1.. ...: .2! J‘...".- .3... 1| ‘3".‘19. a 3’... ... . ... o. ..‘1. col; a. ......- i- 9.0.033’... 0 I‘ll .... cl. 3...... .. 3 .. v.‘.!3l .0 .03. ~ 18:! ...... P t c. 9 D . )‘7' $03.23.. ..I’.. o. Vf’!?‘ ’g’lOtlo .0, ’1’)... .35.... "sa‘5’: .. a... l..0...... ......o... .o . 1. ...-8...... 21.5. ... .i .1. 1 . . o no... .31.. ... ....05..?'... D... ..r.....::. i . , . I... I .. 2 . .9 .. v ... .. . .... .. .. 'osirlvt. c.’o.t.. '1'.- .Il'lr. Q... ... P. 9 I. .‘ 3330...}. .... ...». ...? ‘8‘k00...3’ 3‘ .. . u 354.... .‘D‘O... II... .,:| [32... . vs 0‘. ... ...X‘o.l.lu.t’ -..... . ....o......!.l¢ ... .53. '13.... ,Ql‘..3t-$l,!;..bi n...“ . . .... . 3r~.lt‘...£ so. ..7 0.. ~o..‘! .. I....Mv... illff. .. 0.1. ... l... ..4.. .V .... O... 7). . c. .u .8)..- {...-Fl-.. ... o. .... .. £18.... . £371.. ...I. “......“ Sf’luvl“... v....91’.....3l.¢ o. 01......‘c... ...}.5..V 30.. a. .0...) .53.! . ....2‘... D......-0“."L01....‘Icv13v1k49’. 9".1“. \I."~. cl... III-a.- .1 ’.§O’:O.§l5afi ..4 "n.’>l€.‘!i- ."ohunrx‘JJ .3. .‘JIuoun16C.a.J-l3: o’cJflt-3...’\o60. 3’ .. a 9.. ..-. a .3“- ... .4.... .J...a.o.\ ‘93,.9 pJI)"‘v’ . 5. ¢ inn-ct. .. ...-4...... in... .. “fil’:7.h”.h.~u: V'I‘II’JI‘IovI.QV’LvhosJAHa’Ifi'b-o?’ .’H-in-Oot.‘huvhv'hpgaoah‘u‘n.’ka.h3$I . Li’tnIg bi.‘§_...“.t. 341,53... 9... 35,... I, ...... 035‘ 0.. ... att‘flux C\",. D . _. a .. . X0. .- ..I3: . u . .. . . . . . . v. . .. .. io!!|§....l.§ .. u. . .. .O’I. ... . . 4.! . . .. . . H. . 7 . U. .15 .533I ... o... 9).J.B..¢..3-.$$.v. ». . .nu’KMp 1.9.0....“ rasumvuuo31~51 .. 5.....1... $ 1.. ”hiiltoo“.~’. ...? .... ..l‘l~uo...|4.‘.c.!q.. o ....cp .33‘.Ifl.§u .. .0... _ ot'IU‘oz: o . ... wtllt...!' p! 2.--... 11...!0...I£3..I$‘. §$ . 13.4.5.3 ’loa.vq.}3l I. ... .‘OJ ugh-6bflnviu.g.\10’.b...\“iut.aub 33,3‘fi3lb.‘ I .93).... ...... .. [)7‘... ......2‘.» ......v1. 2123......3 .31'.I. 2 ) ..I..3.fl9w ...I.~...o......). 2.13.2.2... ...-v.8...‘D33. .P.......... 3.391.... . .33.... .11... 0..."... I..-...... ...... trot-1.8.5.! is» Shiv... .1). ..."... It “ 1|. . ...! .. ...? O... 0 . Ju.).m'.1...i.‘..ods.!.ul_LSJAII‘ 4.13.191...» .n 9.. 9.9.20.9...5‘. &§ 5 .5 I..)... I .0... .. . .....3. ..r. 03.0,.) ....I.‘ a ....-. ...... I). 7“]. .v 31...}...Co ‘ial... .1....--438v’,4..t.)!..l.. ..- l......‘!..o... .ngrlo... it Inleflloiié... ...\~.s..‘..¢ .33... ... .. ...-3.917.... .tl’sl.'& .‘Do..ooo..u’g...g: .9109”). I..-H.692! I \Q! .....b... -..... .....I... .35.-.. ...... pl .I’EI 03.8-?3 . . .... 279 .0. .3... 5.4.... S... 10.. ot..|..i..x. ... .. :3, o0)!..d...vt..3. 5...... D... : 3.330..-. .... . ...-2.. ..l...... 11.29.... .3..)!Q.......O\:! LISSSI I..........ib.ll...l. ...is..o3¥‘.9:.. ....O..7\9397!’..§2.. ‘09,..vt Pt . .sun.’ ...! a)... “$.1kui. it" ...). 39$...0‘2. . I 2...... do... . .21'5. Q. .i if . . -..... ..3...‘: 1.4.. .IOJ$.K. -. I... 1-0.4.3... ..l .. 2. V. A... . o ...... I.....fi". c33.£3,39 :03. ...... . .50....x'ls.oo...0o.. «30.1.. ..\ .... ... 1'9 .... c..." 2.0 .0 .n .. ova-......Is c.9...‘!s.7l:’.o a...- .‘o...’. .... 3...: .... .v.“.’. u 3 .‘o..l‘. .6302»... .0...- .. I v'..- DI 4“? } rulllu! ..I'.’ .0.“ .‘I ..Q. 3"... . .....ool . 3". . 3.....3; . .. . L03: . ... ..'.. 0.. . . ..vo .9... I.....“ .....l )‘l ...0. n... .. ..l..“31..~3.«£~.' . . I’ll§.-..o"3..ni .mo’lxt..¢oQ-In9.."..o. .QSI. . ... Q: I33..'o.tb.—u..’l\. I....Ia.t r."s.7s.ll3v..as.i.o..0'.!§ . '3. ' 03". . ‘1 9‘ Lab...“ l‘lukrfi ..‘ra ...... ‘ k ..Iflu nl .‘§!h"§ .0 t. In... 53 .l . a. ....................,ouob.o‘ ......‘3. ... lo 3.21;, ....‘Jstll ... 1.0.3.3.}... ... ...... 03).... ’8...rII..a...-h.a..l:)..:.12... ks. ...}...éql ...9....-.|t.... ....I...v?.,. 7.2....- .‘...... 38.7.. 2...; ...}..I...’ ma“...!~,~..: ..3)».¢9‘-.90..§.2~§'N.IUP§HN.. 1‘ rpfiiirafi vii Iv....7..9 -‘093 . a. .... 0 $2.11! I..}fdrxlfifia. ..b. ......r .5...“. l‘..... ....b..lo...:J.o. .337... “Q .... $13.00 ‘kqofz‘... I. t ’13... 4.......... {...-...-..ri - ‘5. II’.. D..I.. . ...-......38... ..I. 3.1" a . or... ... 1!.i$..o.to|o.2u. . . ... . ... I.1..Ir.v33-.‘..o.zo’\. 13...... 9. =33. ‘ . I. '3 ',.I....h in. , {05.315331 ‘ lh..‘.8)..3.0i-)§O.¢.9 .- . .2"? ‘2‘. $33. vlit.‘ ... .. ... .v . ......Sd .. ....o). ...... ...... .. all}... all 4.6.3.8.. a.......1...:!..J..l......I..... o. I..}: ...l...3.. cl‘...\........ .1..33l€l!33!..$0 ......o ...-0.! 9.0.... . . -... ...... oh... ..., 1.38 ...-721.99.135.32. . ...“. o. 3 (nine... .59 ....Jfilvr ...", b I1... 1......Tllbiace $323.00.... .‘tazbiq 3.1.1.3.... ... .potAo'.-l)..h (.144. .o{.. - Jc.n.:1 .11 . 3‘... I... .....i ......otv ...- 5......3 3’11: It. o3.....fi. ..t. I ‘ ..i...’.v..$“3.2..oo ......i...l. .‘8.i§.....l.. 36).... ’Lii‘ l .... ......9 ..9 .91., .0. ’g)’ I... 3... A "i‘! ...-II!!! t k . h . . 9» 1'.» . III 0,}... 1...)! '1‘...‘ v 0.9 ..I . ..."...q.‘ ...,“ot’ ...... . o... .. .....v' n b -...o '1 3...... 31...... ...-...... é!..3...v V. ...? 3.0! 2.333;... . r . ... ... 3'3... 2:... . . he. .... _ ... ..l116fi..s.3....c4.. t......l....t!..l!0! ....03....!\. (:1... {III-...:9QI: city... .0! IL93IOII m.- .104u..‘..ll.rthn 2.2..1‘9310 O a. l 9.... 32th,. ...-0'; l. . . .13....9 ...... .0 . £1.03 ....‘...‘3‘.‘ .31. ...“. v1.0 .32.... ..‘Ihao'rxv. 5.0! ...d.“ x 7'. .. .0 .3 .. . . . . .s.'.\..lo.u. .2 ....l ltufgv...’ ...... lt.3?$\.§o‘§c~0¢..:3l .‘3‘!.\0!.£"’.h.5. ’ .‘rurllb 9.1!...‘1‘ «L. \...A,\ v“ r .fl‘I. ’- . 29“ .."izt. . .. .1 5.3.3.1...)01. 439:8...1.1.3.3:...393 .v 60 1.1.2.3.}!!! .... J.- .....I.. ........a..... .90J........ 3 .. ... .1283... ‘5”. (...). It! ......19.‘ ....tof‘ .....P. 9..”3'...I..‘3 11...... p2! . ..., .. .ififiip... 993;’a3..§"\.b .3}. O. . 1&3, DID..- ...-0., .. ......3". ~..t...flll.. ..J..I-..Oo ill... .. I" ... 0.0... ...b...‘....cvo.o. 4.10. l....o.v.¢§-8lp..b) .h..é!l\u~.;...nv..fio9.l” . . . .. c . ......1 ‘4..|’ w c. ...; .12... 1.0.3.. -.4.?)oofi 0.....03 .. ,..oI-.olbtvoa:. :J to. )3 ‘Q... L!..¢O.3.. o ...-L.‘ .1 1.0.1.!» If} I . 09.19- ... 4...?! .... . ...)... ... 8 .u 956..)Ditdllo ... .0703. .. ... ( 1:3). I..] I. o L. n i 1! P14 ..8‘ . .... .I‘o. ......u....l.. l..oi.....o..:. I. .43....3...o ..va..-§34 .‘&......4 .. ...-lit .... . ,1!....)Itzloa..ri , .. .o 3.]... a. .....o» .J..?.l.vo..lo.v.o.n,... ‘t....).v.. I Plv.7-1. .. .Dvn’.. ... ‘f'33.-33..!'| 6.01:..‘1... .0},...¢..... ...-...“. . .....IT ...I 1911...: OC.... ... .0 53!)... . ...... .33....6KV: It: 233...... 0.. hos ...-O. 3.3.... ...)... 3370...? 8...... .. . ....CI . ..L......52!2?..I .1315: .0... ...?t..12.. . . . it - ..L ......o .333... 1......1Jv......30l9.. I..»: .. 3.41.1. .....v\3 .....x: 14.51.!!! 1.3.30.1... 3:22.... a) .1503... '31}....II..I $539. 4- .....[3 ......a «3.3.3.. M.vfl.0...u4.n!9..o‘.o ....JI'QJ.J.... ...". 9.13. 9 o ...... g . .o a h o. n . . I. ..‘1. .0 . .o o... ......30. . O . n u to pt. I t... to . I: .C‘... .. . v‘ ...O.... . . . I’UJ ,4. .‘IOJ‘..' ...-“.. .I.’J-.~I .. VIII. .Q. o r! I ... ... )‘ .....Q‘s... 3133...... .. ...... ... \ .3 :34. . .... ...-o.“ .....o...’.oo. . ....lum,...ol.r “.01.... .. LP”..- .43.“..Jmuo‘. ..l..:....\.’.w.3...3'.. ...-nonunitiv tau; 1.793.... 2.2333‘30... 0.6. A....)~.9!?...?~.fi... ...DSI.*.“- . 43.0 UHF... ..hshlkhoo .... ..‘oiov‘ l a. ....o 33.3.3.3... 793:... 5.2251...- .. 4.. .8... .3... . 10.. 7.3.... .b. . 05...... .... ...... #2065 00.9.. ...)..321313... ... v ...! .... 8..'..o§o. I o. ...!!! .....IO..3.¢.8 2.213.21M! 3‘52)! ll . . , o. in?! ...... . . 3.30.... 3.0399... . .... ... .. .ltxstt . ., )....v.3.0.........cl.. .. .03....14. ...l....§. u!....9.. ... ... ... . 7.137.... .I .9... ...!7103 ...-...O‘Sf. ...: I...“ .4 . ......v kt......\(.!...t......)..$.. 263.... ..- 3... .- ii‘..t ......"- . I.....QO... I, .... - I!!! .....o. ......I‘U...3.A 9.6.1.4.... , . a: .... . l3...p.2.u1..§.l3.o3.v . .v!‘ 9.. Hz. a. . .22)....2? 3.2231! 193.... 031-. ...! ...!!! - ......" _ .4.. .0: ; 0"... .68...- ... . I.. . 0‘... ....4 ... .... ..I‘i‘l... Douooé‘O...-.O . .r. 23'. 3..- 1.4- ...7’1I1. .... I11. 9oq~.n.7.o qcs.‘$o.l’.. ..., ...lt.oao .Alo‘ {I .0 '3)... I...‘ ... v ......4..’. o...) 1:03.... . . .. vill‘t..‘3a’..l ... .11 .....I... .o!... 21...... ...... .07).. . I r...‘ . .0“. .. . c.....<.).. 3...!!! 29.... 6... .3: u . ... ......! . ........5illt.rl.l...lo.~“§f..t7 .31..)0 ......1... b... 29...... ID3‘353333u15Q’ ... I... . . . . v 1 .f I.. 3...... . . ...—231...; .ll.. 1:08.03? ’3 .0. . s...4(.v."....-:... .v’ .. .... O 3....t....\vb..l-lcnt ...O._.-’.n!!.ll..a .....- .I .9 . .. . . y.‘.‘\. t..- .J... 09.-...0.‘ ......X...‘_..o..vl.. 01-0.. ..56..‘l...vl!0.. .. l933301..1?:.. Q}... ’..Q..!.i..3..s.1.y..o~.3.60"»...1 obol. . a . \35.’.‘I§Il ,l2. 0 'tor.’ .quflollttp: ...-a 0.. 5.. .. VA. .5..‘o~c.. V .07, 3' “.Jv a... . ... 3‘35 .... I... .. 'oo'..£0'. ... to. .U...II.~ Coysc‘z. 302...! I; .‘P‘!’... On. . . . ., . . J ....“0’r 01" vi. 5..‘ ‘.. 5...}.-. q . . . .33.... o..‘.0‘..‘3lfl....=\a..7 a... O... J" JQ...Q. .u.’1."’7).‘~.§.“0.0:f"z..$’§--.‘. .0 ...." .0. Q l.'5.0. 1, . _ . . . . . . .. B ...!“- . .... .lS...‘ , . . 73...... .... ".og:...\...l.1....)’1'onl’.t... ..O..7-t.‘...l..5o.c.glvo\J....t\9..l.1.\b (..4- . . . . . . . . ..‘tvz. . . . . .00..a,.... .. 3......32311232 ...)..ly. t7:«.3:..:3§‘!.1.. 2E0! .... . . . . , .\.gs.l.~:! Aulfilso ...)...u .3331... 43...: 7.. ....V... ..92...Ilno. 8’....I..o.t, .... ....."1 u ..I'. . . . 5. ... .D.”aa. 19.8...- .P‘o 9333..‘z.. It... ..I ...v .. I... I ..§.I\., 4-.‘Oz.’ .3...1b.‘..t. ....) ..3 .... 3‘3‘1 . u... . . ....no‘ . ......(:..\........38.318.71.32 .... N. 2.2.8.889: ..u..vl.o.....o..x 7hr...:.......:.t.n..ufi. ‘I .I 3......" .. ... :5 o ...... s a. ’3” 0.1. ... o... .0 CA 9 figs‘ ‘. I. t . . i :1 a0.“ 0.). ‘ .4....8! : 5.211.... P.Pl...o..l..~u..ohn- . cl . . .0. ... .. . .....WMI .6... a. 09.0.80! 19.011.135.11; 3. .>..v'3$\$v 11...)... . . | u. I) ...K I... .1 " .51-..3$I§N..8m1‘5. <8 ..c ... . . ..1 .3..‘.Dnfil.v , ..’C&3..:Jf..":l, .... ‘I’fl‘.:."’!\n.’: ..r!’~.'l|¢',§‘1“".o!I....‘vnl.§\. a.VI..n .0: .0 Q. 90.. . -. ‘z..‘~-§\'.¢“§ .. . .. . . . u. ti!!..3.o .... 3......)3 ... 13...! 1...... 31.9.. 2". 2.1.. b......\.|.’.o. ......)O.....r 3':-..3..’....‘.’.:10.30.....a‘ .v O....!.3\§L.I)..OO . A s I. 4 LEVI .Dn-‘Ilr. .Wfi? r11 K r0... . . o ‘3'! 3..., ‘9. ...: I..-$5.. 1‘5... ... ......o. 10......0': cult-33.. ...: ....uurf’ihain’lbfl." ..JLcloalgl-c ...)... .uoflll"."&§3hodlrl".uaofluu....“...‘IIQPVQ Q... 6.“... ..‘901: .. . . . . 110,333 .‘k’ . . . . t. . n. . . . ‘0' «I I: vuuou .l- . .o . I . . . i, ' v . .... -.. . a. .35.. . .. _. ..3....o...nll.4un..o.n.>..du.).. s «...... ....§.0.99....:... L)61\.€.I.... .39....0: ..v ...... ..., ..l ...... ..\ . 9-.. .......&)Dd.’.| ..WUQti‘... 3.3.0.13..." . . .. alt ‘3. (.0585 .....v. . 0 On. ‘3’ Or. a n! ... f... ..v. ...-0.. ..vv. uo'.‘.)< .05).. ...... .0. .-O§-J3§.X.S\O. no".nhp'l...$nncl‘.~4ho.ll..s..¢b’€‘.90..” o-..-.‘..I".‘o.n.svo43ofio.u‘ot. ..J”vu.»u".lfl"url\vlh“fla ...I..o‘~o'$ .0 . ... . .55'. '0: 0.5 . _ . . . . . . 1‘ ... . . . . .. ., . ' ... : . . . I... ..o a 1...... .53....1330 ..vlhuflwo ...!) :20... 3...: .L....p...........la ......7...!..J>I....-..r...ot.. .~ .I1l3.0$..‘..§¥.§0. . . . .....rtlgkvx )l. . . 5'30!!! 92."..‘33. I17...- .Q V§Q.I1..‘ou...'\‘ *9... "al”."i3| Ii 5.... .15... 4.. ..4.... a... ...2 ..ll....“\.! 1’ ‘33.! 4 .. 313‘. . \.‘$I§O-. oil-.....-- 3“. 0971“... . ... .... . . ... . v .. Al. O ...-1... ... g‘SQ‘. ...!“16 “.... 3%}: 0.0.1.. . . 363......» ...... o. .38....70351 ..“V . 03...... J.‘ ..Jorihq.“3\un.f.o 3...... 3L?.Joo.v.“ , o... . . ...... .\ O \ A ...-.4. v 3'; .Illtl‘AvO‘L 1 0:6 I... c: ..l: . v a . . I No. 39...”... has.)...... .. ......zkai...a..fv . 4.3.5....toiilu .. ¢ .... .......r....r9 80.... ..6... s. 708.100.1703“. ...1 I.l. ...-o2}... ...)... .. .I.. . . 4 7......) ’27.... . .9.- Oncb‘).....o .. o ... fi .Ol. .Il s... . . . . . -... . . 3.13;, .....l.’)" .“.A‘O‘..lo ~l‘i.l‘h‘..3:zt‘..3?}o“‘l.a‘§os . . . .J... .. a)..." U ..uflmhfi ...“...“d. . cu. ... ..vk..t...L..7 . . 3...“... .. .fl.!.v........u..h..t.1..t..t oli......:...n...uu..pu ...s... .u...... I. ....I....u..u.$~ufln.....| “Woxoflwflusomr . hfiflh.u.o.ob:v..0.uhv.. 0995.0..r.t:..!3.0> .- Qiitl‘)! . I.....n’t?l\ .o ‘3‘... . .. . a .. I ... . . a I . a . . . :..3..’ Shot-...! . .... it‘s... :12... $..§ ..I)‘ .... . . S I; .9. 4. )2 . .039 5H! 3.5!. ob“t...33..‘.o‘.. 9" . .....7. . 0’ ... Akitghlt. J."s5htl.¢.uorll.vm....:tb....iinn-...TNWOHIJ. ... v . u...vs....¥’o.9’v.t!.v1..$...... . ..‘2 5:09:19. luptoutéli..l¢ll{v ..21":&‘go \bt. . . ¥ .. . ... . I.. -’..'~.OOQ..,':9 i .Jt. . .v...u..v w... ‘3‘ ’3}. b. .... .aw. . . 6-4012130... .".1..I.’b.uv...an.l!o duvet... -l)’ u )31 ’53....t‘ \.. ...Cn\o..u\ o. ‘1‘..." .. $3. .I . . ... o .......l .-.O... . 19...... 0. ..s I.....v......9-....._..§......,. ......fi) ..--.J ...”)...1 Jfifiofi.“ .ao.w.fl.lhuw..1§£o§ who“... ...)‘Crdflhis I?! .1910. I.‘ . . -..-.94.... - ... 1:33....)5:..RSA.‘..‘II,‘I..! 4’. a} 'u‘.’ . 0......1. 1,....¢.o .II .9‘.’.?t.k ... .... _’.I)‘!z\. t I.. Av... _ . . ...! l.- 3.... ....» 03s 36......3..1l....\.. ...... . ....Juukuu .....1 ...It“.nh..1fififla ....” ....- ‘gunuublqilsohhflnuv .I-.Hl..3o.. I..I.1§.£. .. I: Itfitibti‘593 ..‘c‘btli’x I 5:! .....o: . I}... n .‘.\!~.33.!2§olon.n DE. .123! £n...‘?l..§¢93’2 ...??? . . l¢:’\‘l.. .4.. 3’3)..." I..}. t ......fi. . ... O . Ida... :..o ...... .l)......t..3?. ...... 131$G . ... l.‘t‘.¢9§)t:. Div-......vfltst: .‘4 ... zgl. ...!‘i V. ...... o\\flhifi..oi . ......1 o viz...r1..l.....!..1.al.r. ...... .. . 3.... 1.9..-. .335...- ozkob... ....“ ..I... Jar... ikfibfluualfllooufiu...fl.rdcfloi awhtdwgfisifl...ah.$kkr Jagvzfat‘ ......)9:¢..!9§ 3.1.3.1.. ... $332,)... 3 v. . 1 3! . 3‘ V: . . ‘ l.)- . n 1 . 1 l . .I.’ o. ’1... 0...... 'a I! . '5‘ 2 2 ..4}: ...-...! s. .I -31: ‘1 . .. . . -I.K:..O..I.. A... )gr.unr‘3ltr.1fig’ :2 Dhl\12sflo- ,0..1"...I.§..‘917).Qh.031§.1 "flfl Ti’Jvcc‘hnfih‘ up I. ‘o—il.‘b~1’<. - .a‘n... J. in! ‘1‘.‘.|.‘.rt"t. Quozoot. I! .... {.2 ...... .431... . .. . ....l no. ’,I.\,:’ I)‘ I. fi. 4. .... . .... ...??h... .‘UI’: ho’t, Iv...-zlfi,&.‘3!3.nu.|‘32.2.9.0: '1... lb Q... ‘.I.‘ ran.“ {.1‘03. ‘1'3’305, 2...... .....l: ... .00... ...)..2: .31 ... I .. ... t... .KR“..:D:..OII378531!13. §$lto.: ...-.51.?! ......)fi”..- n...1....§.DDV) . iiSI. .....x...l..$!b I... 20’... . fix . ...) .a .l D .1 . ... . L .33... s~.h..)..|......1§ ...! I..}. ... ...... 32"... ...... ...!)..§.~.DI..I..§....I02...1 flfiufleb‘lt. ID; - 21....)957. ..313t3\!$i ‘5 . OI .... ...... . . . . . ... ....X. 1.00:1... ..1..9LOS?J3.€.’I..Q...¢?.\xus.‘.. 2.....t...\\.2!01..31iv 2.1....“ ..I >0"’b!lf‘~o [y r IIJ. .n. )I‘Isti’5$\l9. ......z . 4 . . . . . . . .. .. . .. . . . . , . .. . . . 4 . . .D 2.3%....339 ...)..th yisZ‘io...?tl‘)..l...vlsrf.91:33:»! 1‘... 3;.1‘0..3..2!‘$.. . 1 ‘ . - a... 0" fi {“52} .3...) ....4 . ...... .. .....r . , 5...}... .. . . . . . . , . _ . . . ._ . . . . . . . 4 ., ..32....1............lhu..:........s.......u....:....s.c....fi.vx.“n3¢...v h...!..§1.....ufi...n........ 3......“ ......21:x.........:....€!. ...v...$l...¢....2 .3n.............. «3......» .. ...-o ... :1; c; . ..T. ....) t .23... 7...... . . . . . : , . . . '7‘. 0...} ...... .0. a. ....I .4 . \v. 5.- . I. Q‘ U .z . .1!!- . . . . . OI . . Iv . .15 . . ..o o.) } .5... S.\ O‘S‘ o\§s§.;‘ I 0 . ‘3‘! D... 3 w... 33......3. .43..................._..v....... . . . . . . . . ...... «Fianna»... 1...“... ..um...u.:.........v....uflvu .ifiu.§s.h.nuhlsau...nfl.§ Shaiwutd ...... e......~..s.>..4..ca9.u....... ...oviznh.£..gih «Voila»... {rung I. . . . .. . . . . . . ......vl.... . . .. .. . . . . . 9914.....05... 0.1.. D .‘9 . . 1 . . (I... .... .. .0 . ..Io.....l . .. .. .. . . . . . , _ ... .....\ . . ... .... .... i... "39..‘| 3 .93.... I... o.- ..1’..0o 33...... ...... .9\ . i" ...... $9....)Iosi ... ..\Dl..... 2:1; 3...... ‘18.-.. O . ...-w... . .... b. . .... , . . . . a. 3.1.1.... ...... knit. .r~V§..a..1..¢.1932.c... ......‘o‘t... ......,I ...-3. I.. .‘..9‘....‘.fl"‘\!t . 35. D 4.. ..fi‘ «...-90“... z . fir o? ... "OI .3...‘r.olu.| m ....u... .h..... 1‘. u . . . .. .. .. .a of 9.. ....uficun .9 ...“... Juno... )3... .‘0’:wakikfljuufitJ......thflmnn‘.h.):hu.&.«.§ionunnxasnh3t~‘!)«J‘I “3.4“. R2. s....s.vi))§dl¢‘s...- I a.» . . . . . ’ Q 0" :1.J.o ..A‘!.” to .1 ‘.| .... . .‘guiu. .‘0 I‘D . .. ......»a...’ a .... 2.: ..- ....ra..558 ”ungiflquuuhqhqo‘hfir.‘hunvlowuohluuctfi..nutt.u...o.l...u.§fiD .99)\.0...I))D‘| ‘2 ......‘T. 53, "Olhil ‘2’. c. .1. I‘Qgflngn ”*3“ 49“.... .33... .S‘AI.‘...I. ...: .Q 5.3.0.. .oz,‘.ou.o..C.)R-1‘.". .vqifldawrfihwv’. .- . . 'Lfl .30. bk..tr\dfl-HM¢.‘yzflgVXMOda’vWQV‘OW. a\.hhvtl Q ':.\..o. ... I..} ‘1‘“.70 .‘ o . . u .. . 4.. . I .4.. -..1 ... ... .... . ...... ..«flzwku ... 1310”"..0859... bus... I ‘Comutsn.§tl.ub$” !m\u‘g§,§&..7.o.:. ...)|I.Q\.9! 533‘. ...Q ;.j\hoz......! 39531.5... is ... oo. 0 0...... to...” . . . . . % l I. 1h .1! I .- 0.1.5....)‘rl‘ ,0 .f.‘;};lv. .‘..%!.‘g o. ... ’9“i‘g} ‘3’. O".ll§\ .‘n” P In. . ob ... . . .... .’..1.9!.I..I...‘.3..r.o.0 .... ......l ...,igtiltixtt 3-19.110.‘0...851....J3..Q.1.. Q$.§tb...o.5.tlc.l‘l§.’..§..1¢l tic-IOTi .3. ll .. . g. . . . . . .. . , .. . . ... . . . . . . . . . . .. _ . . . f... .§1.’....I..I.oo.l.. ...... b...” .5.31?.l."i9.ll . . Dart). ... 7"... 7 “.l ...”!OIH".“Q.‘;\~95'”W\ nun.ils..'5| I." INN": .... 0!... . .. . . . . . . . . . . . . .. . . u . .. , . . . . . . .. . . . . . . .....S 31.... ... .5. ... ’83.. v.1....)s.¢. .03. .... fi . i 0.0 . ..-, . 4 ...... II E h. _ .... ....Hmo A..." are. . . . . . n . . .. .. . . . ., . . . . . . . . . . . .. u. . . ....‘l.$.uo.<...0v . ‘00.}..’.t lu’- . £\\.c"."l¢.“it‘ ...” .il‘o.) C... '3‘ A.::’.’ '3 .. . .... 0...... . . . ...: .. I3.... . . ......«3‘... . .. . _o...OU .lb..,......f. . . . ......f .....h .... s-.. . .. . .. . . . .. .. V. . I C 0 .v . . .. . . . ‘hflh. 62.48.. J) . . . . V .. . I..-o". .. . 0 ..V . . .. .. . i... in... . . . ~ ‘. . o“ “... ...u... .N o . . . 0.7.0:-.. 33.3.”... u <- .. 120...- .. ... . . 5...... .7 ... . . . . v : . . . . ....- ...:v0 .. .. . . .... « . 3! I... .33.)... L.b...oils.§0!lt\9‘i.3...‘:c 9.3."... . ....u... ... ..- ......w........._.._._.. _ . , . . _ ..-...d...» ...... . _. ; - . ....u...».$.?.........1.a .....I.........a...,a..............u.:..........s. ......9...n2....523.. .4-..».fi...?...b..l.t....u. s haw. .. {finifittsofi 9:... O. .- ... .3 .. Abbi} A. .. . ..v 5... ’5'... ‘1. ......) vb. ..-...c) . ....Olclc .t... IOU..."3|I&: ’3! .’ O ‘3’. .l. ...... I Q. .... in. . , . . . . . IO. . . u . .. .\ -.. a . ‘ 0.0 .... z$l.'.‘ J! 9’. 3.x!!! 0...... 6 a 0.5... 1"..9!. str“-.. ‘ .0. .0 '1‘". 3‘ 05.5 .. A... .. . IL [H.WQ‘flm‘ananduuwu’} .‘kfinzldsoo'uv‘..:t39" “.ngh‘..ov§l.0.ltlmua .4 a’fi‘o.ih. “:." ‘i...l - '0‘} ......O... ‘1’. ..\a..t—'t. ...... nu.)- . . - . -_ . . . .31.". .. tiugla .....us...afi\.....o.9\i 8.3 .. .3 3 l5}... . .-.-...)..vluofi a..3$.l.‘§§..l7§?i|a 3 lo :IUSI‘.9 ....6... .... 1 I...lb...3. ...) ... 0‘85" .... ....Ox. I«d....3l:.o.......o...'~ o. 3...... a: do I. O . m. .. .. .. .l . .0; . . - 4o... P’D’LIE: . . ’3 . ~ I .. 1.. 1 5,2-.. ....)IOté 017.51 0.3L“I.5 92.. 3.1:}..g'...’ “cu... .. . ....u... 0... ... . o. ... 0.. po. 3 . I .5." 3’12"“.3b unfflccahnn of:.\...u_. it! (I 0“...” \ (c“.\. .3\:;’! O\..O‘t2£\‘g ggtufi“ ... ’10..._... .5... ...-...»; .... ...... A .. .. 9o ... . . ‘5... .Q n "“3 ......»s. ”1.3....“hzév ‘...pl.....bl . . .. . . .k').o:\...... ..‘Q...S§q v-\.. ‘4. v.1! loot-o . .l- o s I ~ ‘9 . . . , "yfillfw I § 0. ... ~90. . . . .0. ..., . . . . n .f I . I . . .0 .... v r I u. . . . . . . l . . . ¢ . n. . ... . .. . . . . :3; . .. . . ”Prawn... .... ...... 57.52310: .. ”a (t... 0.... ..«......vu... can”. . 373.....- .. a..:1x'ob..... ...“..itfiblcl‘xl; A. . .. ...-u... . .98... .vtshuudzavzn stun}. 1N0........¢.JO.. 1.1.? ... . ...}...‘u‘ in“ 0:1 . .. if... .. .v . a ....,...t!... . .....I. . L». . an.-.) Z. . A\.: ... 3.13.3 ... ......"L.o.v....9 .... "......‘Da... . 1,...T....un.)....aun... fifl:£fifi.:fl.wfl. 5.4.0.“...m6. 13.3.1). ......lc... ’ 5)... ...Qll. lo b... 5’1~§ fi t. .V .. . . . . . . . . . I . . .- ! . . u I l .3. o u ; g . 72". v A.“ '2. , I & I. l’s‘. I 4 ... _ . .. .uf.............uv..._.a.!.rz.u4...»«.u ......... nvuuafiflhud 5%.... t... .. .... . . ...I: ...... . .. . .r... .33... ..1?...I..s..‘)§... 3 .....zts... ...... ...... ‘19v .ioz 21...... ..I ...-c. n I.. ......IéLvfl: 3. a. do}... ... ......f..t.}. ,. I... ......nlht 3... \ '0 . . . . {ti-o. i! ¢. ....d . 0 .~ . ......1? . 1 .01.. ...v ...-1.03,: ’ .... ... . ...... 0.1.3.... 3. a .V a. ll.‘ ...nI. «1 Qt... . . . “t’ll..'|”ui ..a. . . :I‘Vi. I "J o. . . . o . 00“Ilb\(.b.)|“l . . . . . {g} . ' .. .. osoz’b I l. "z‘. . . ...) ..oo. . g -..-d... U... .0.- A . O .. .... ... . ......tV1: . E .... .. . . $0.01...-) . .. ...... 5 .1... .. 113‘ . .. .... b . .... .. . . . ”v I.- ....th .J .SJNHH... v4 . R.- 5 ... . ....o I. ‘0. .9.‘ ... p..n.¢ .033...§t .. L’oo. . . . . u . . ‘ I.. n .’-I .. .F. . Q o {.9. .v..‘.a.....o..... ! .,.C.....d.... . . ., ... A, . , . w .U . . .0. i...“ ‘0 U. . . . I. n a n . . . , . . . . . . . .0? .. .. ..4.... ... h. .33... 0L". . . . . $-0HR-MH .o.....uu....).. ...: Hath} . .I.‘g . VI... 25"...“ ..l.‘ .92... .. a.“ ’44?\‘ . I .. ....- » 3c. .- ..“. .v‘H u..r.\ -T .. u ..a . .. . . . 5.5.9.4 ...: .......o 3.3 .u .- .. 'c. .3-..“ .‘I a.“ ’.l}..w“o..l.lit bun.- ..” ... \’.MN\H ...: ‘ounn.nl‘,..i§.¢§2.o.‘ Ito- ... . J a a . . .05.... 13.3.5... ...I... 1...:- I.’...6..I. Mu... ‘uhuuiu .01.“. «ume‘NUM. ...‘u’fl..l0.5"m- l\.. I On... ...-"...V~..)o . . . L .. In. a... .ovir.. C .lul. ... ...: ..o v...- ... ..... . . . .. 0...:laolt'i '5 2'7 no... .A Q... . p . . .‘I. l3§ £3I‘)‘,zn 1“? .‘§.‘\.t ‘.o ‘ v9.3.3; . 6“ y. .. ... 3’3...“ - . . “...-9| 6. .. . . . ... . . . . . . . ....S’.I'a ’,~ z.‘ l0....l\v9.to" 2.5.3, .3"... f“ 03?.“ . . ... 20.! a . . . , h....ol..n o5. .. o 3"..-- J. . 2,, 0.9.3.6.. . J . "n .7 . ... t o.:‘.. ‘~.ul\..¢o .. . .. .0. ...; 2... 3.0.4.9..35l6. 1.. . . . . . I! ..v ... . .. “n” V’,‘ ... ... I». .. a . ... .... . . . . . ...... 2.3....3‘ ... 11.5 : .0 ......si...... 2.... .‘v. 531.... a... 5......" .0: . ...... L11: ...... I. . so. 0 . x 5...}... , .. I.. I..} r .....-‘u... v .. . .. 1.6.0.2.). .3 .I. ...... 2...... ..N. ... 1.3:... a k... 3...»... ......,.)_o.’...n\ .... ...... . ... .3... f... ...-41.. ..I.v.....s....,.. .\.§ .. .. .. ...... ....4. T ..s. . ..- 1.5.6»... was... 3:5. ..-. ..u: . an! r. t... .u... ...... , 4. g .. ....i..- . 3...: - 73.4.... $3-... ... . ..u. ......s....!............. . t . ..- .. c .. ... . ... .. ... . . .o . . p . . l . ..Ii . .. . 8 .. .121?! . . ........I.I.cl . .0 ..z’....’ I n .. o. . ...I 1| 40.. .| n. I. O ..O‘h..’\. 5.0.0... . ..s....o.....‘o 3.3.21... . ...v1.’.04..0..i.vl. .9... ....v‘ . c I. u .. .‘O..ac..a...4v¢:.’ .luvtfi ‘3‘ . .-..o... .... 0.4.73... ......filr 2." 0.) . . i . ..D .. . ....lc. .- 4 9 3... 1.....‘.\ . ......ué. 3.0.3.00... .... ..... ...; ... 3.3.2.290. I..)..oul‘. . I . \o.‘.o H . v - 0’. ...}... . . .. ......3........( n .. 1. .. it 41...... ...o\ 13.3....‘h... ‘ 3’33. . l...... V 5‘... 233000-02... I: .V)?:|v . I . Y5 9‘ 0:91.... . 1743...: . . .. .V .vul.. ...: . ... . .i.‘o&v‘..‘. :12...‘ o...2...h..l 3:... v. . 3.... 8....- II 1.95... I. 3;]! u . 933M......,3..‘Muo.1.\..9..pol.‘ ... ...). ..uov . n \ . . . ... ... .. ... . .... . I6. ...... .3... ..u. .. ... .. 4 3.1.0.3931 . ~\........1.v..0 . ......1. o. . .fi .... I. ~ 3.‘ 1"... 11.....1: '5‘......11 3‘3 .... .... . .. I.. . 5. a .0..- . .11..- . .. .\ n. . 9.. a. .9. ...I... I . . IQ... .\ ...: .. .. . 2.0.... r .....50 .0... .... ”.~.o..'nlCDI-p.c 0.! 3|. .. . ‘9 l . 3‘ . . k b ..‘o. 5 ..v' . I;.’. VG. ...o 1‘ I, \n......... ..I...I.tls..a nil-o 1‘ \‘.‘. 0 ...~ 0 ‘ .‘ 5 0‘ .‘ or... vi......v..90.o-13......E...‘ . ...V. :3... .... . ... ..0 .. 4|. ... o I.. o . . . .0... II c. n c . . I , ... . figggaé I . V . . . .... . - I n . .0 r a . 1.. . I.. . u. y‘.. \. ... n n. a A ) Wm... a. . . 3...... rig....1z..u... .5. agitahwmmflc .....fixwgfi, . _ 4....» . ...: ....Efigogg... .m..nmv§T.. ..am..wm..a..uuhuu.m..6u. “gawk...” Wham...“ “mag mgr. . 9 . . .. . a. .. ...-...... LI"... 9‘“... 5. xlwu ...: .....O..H. 3.. ..o ...... ‘I, .u.‘ .1 Ju‘n}... no... - . ..Iohfl‘49: 3.04. A .nt?‘1.... . ... n I . . . . n . llllllllllllIllllllllllllllllllllllllllllllllllllllllllll 301016 4667 This is to certify that the dissertation entitled Deceptive Message Production in High and Low Self-Monitors: The Impact of Directed Versus Undirected Rehearsal Time presented by Linda Jean Marshall has been accepted towards fulfillment of the requirements for Ph.D. degree in Communication Zmfl/ 2.6: Major professor Date October 27, 1994 MSU is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Mlchlgan State Unlverslty PLACE ll RETURN Boxmmwommmmm yum To AVOID FINES Mum on or baton duo duo. DATE DUE DATE DUE DATE DUE MSU IoAnNflrmdlvo ActionlEqnl Opponmfly III-mulch W1 DECEPTIVE MESSAGE PRODUCTION IN HIGH AND LOW SELF-MONITORS: THE IMPACT OF DIRECTED VERSUS UNDIRECTED REHEARSAL TIME By Linda Jean Marshall A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1994 ABSTRACT DECEPTIVE MESSAGE PRODUCTION IN HIGH AND LOW SELF-MONITORS: THE IMPACT OF DIRECTED VERSUS UNDIRECTED REHEARSAL TIME By Linda Jean Marshall This dissertation presents the development and testing of several potential plan-based explanations for the apparent deceptive superiority of high self-monitors under rehearsal conditions. Experimental manipulations were also used to ascertain whether this relative superiority could be eliminated through the provision of relevant information during . directed rehearsal time. Study 1 assessed the level of self-monitoring, degree of past experience with deception, level of goal desire, and desire to plan both generally and in the specific context of a hypothetical deceptive encounter. Results revealed no differences between high and low self-monitors on all outcome measures except desire to plan for the deceptive encounter. Individuals scoring low in Ability to Modify Self Presentation (AMSP) and either very low or moderately high in Sensitivity to the Expressive Behavior of Others (SEBO) indicated significantly higher desire to plan for the hypothetical encounter. Study 2 compared the extent to which high and low self-monitors were correctly identified in each of five conditions: (a) telling the truth without rehearsal, (b) telling the truth with undirected rehearsal, (c) lying without rehearsal, (d) lying with undirected rehearsal, and (e) lying with directed rehearsal. The data were inconsistent with prior research, failing to demonstrate the superiority of high self-monitors after undirected rehearsal. However, liars high in AMSP and low in SEBO were most likely to escape detection. Tests of plan-based accounts using nonverbal and verbal cues were unable to account for the superiority of this group. Explanations are offered for both the failure to replicate previous findings and the apparent inadequacy of plan-based accounts for explaining the success or failure of certain subgroups of deceivers. Proposed fiJture directions include the assessment of plan content and message behavior as well as the consideration of time sampling issues in the coding of nonverbal and verbal cues. ACKNOWLEDGEMENTS I would like to acknowledge the support and inspiration provided by Michael and Shannon Cruz, the guidance and faith shown by James Stiff, and the valuable insights provided by my committee members: Steven Wilson, Sandi Smith, and Ann Baumgardner. I would also like to express my appreciation for the long-term support and wisdom of my chair, Franklin Boster. TABLE OF CONTENTS LIST OF TABLES ........................................................................................................ vii INTRODUCTION ........................................................................................................... 1 LITERATURE REVIEW ................................................................................................ 2 Self-Monitoring and Deception ............................................................................... 2 Planning in Deceptive Situations .......................................................................... ...4 Plan-based Explanations of Self-Monitoring Efl‘ects ................................................ 6 Experiences and Desires ................................................................................ 6 Multiple Goals ............................................................................................... 8 Plan and Behavior Modification ..................................................................... 9 Distinguishing Plan-Based Accounts ..................................................................... 10 Exploring A Potential Confound ........................................................................... 16 STUDY 1 ...................................................................................................................... 19 Method ................................................................................................................ 19 Overview of Design ..................................................................................... 19 Participants ................................................................................................. 19 Procedures .................................................................................................. 19 Materials and Measures ............................................................................... 19 vi Results ................................................................................................................. 21 Quality of Measures .................................................................................... 21 Past Experience and Self-Monitoring ........................................................... 22 Goal Desire and Self-Monitoring ................................................................. 22 Planning Desire and Self-Monitoring ........................................................... 22 Discussion ............................................................................................................ 23 STUDY 2 ...................................................................................................................... 26 Method ................................................................................................................ 26 Overview of Design ..................................................................................... 26 Participants ................................................................................................ q. 26 Procedures .................................................................................................. 26 Materials and Measures ............................................................................... 28 Results ................................................................................................................. 32 Quality of Measures .................................................................................... 32 Replication of Previous Findings .................................................................. 34 Tests of Planning Explanations .................................................................... 45 On-line Modification ................................................................................... 51 A Look at Objective Self-Awareness ........................................................... 52 Factors Contributing to Successful Message Production .............................. 54 Discussion ............................................................................................................ 57 Replication of Prior Research ...................................................................... 57 Adequacy of Plan-Based Accounts .............................................................. 64 vii Other Findings ............................................................................................. 66 CONCLUSION ............................................................................................................. 69 APPENDICES .............................................................................................................. 70 Appendix A: Measures Developed For Use In Study 1 ......................................... 70 Appendix B: Sample Questions Provided for Interviewers ................................... 71 Appendix C: Instructions to Participants in Study 2 ............................................. 72 Appendix D: Measures Developed for Use in Study 2 .......................................... 74 LIST OF REFERENCES ............................................................................................... 76 NOTES ......................................................................................................................... 8 l 10. 11. 12. 13. 14. 15. LIST OF TABLES Definitions of Verbal and Nonverbal Behaviors ..................................................... 31 Impact of Self-Monitoring on Detection Accuracy ................................................ 36 Impact of Self-Monitoring on Arousal in Rehearsal Conditions ............................. 37 Impact of Self-Monitoring on Interviewee Preparation .......................................... 39 Impact of Self-Monitoring, Truth-Telling, and Deception on Interviewee Preparation ........................................................................................................... 41 Cue Consistency for High and Low Self-Monitors When Lying and Telling the Truth ...................................... . ............................................................................. 43 Cue Production in High and Low Self-Monitors During Deception ...................... A. 43 Impact of Rehearsal and Self—Monitoring on Length of Adaptor Performance ....... 45 Planning Evaluations for Self-Monitoring Subgroups ............................................ 48 Nonverbal Cue Performance of Self-Monitoring Subgroups .................................. 49 Cue Production of Successfiil and Unsuccessfill Deceivers Across Rehearsal Conditions ............................................................................................................ 50 Adaptor Durations for Self-Monitors Across Rehearsal Conditions ....................... 51 Correlations Between Self-Monitoring Dimensions and Objective Self-Awareness 54 Discriminant Function Coefficients, Structure Matrix, and Changes in A ............... 56 Variables Discriminating Between Successfirl and Unsuccessful Liars and Truth- Tellers .................................................................................................................. 57 viii INTRODUCTION According to Snyder (1974, 1979), the high self-monitor is a person who is highly concerned with self-presentation and who carefully monitors situations in search of cues that signal the behaviors that others would see as appropriate. The high self-monitor then uses these cues to modify his or her own behavior to conform to these expectations. Deception researchers have looked at the ability of high self-monitors to use this modification of self-presentation to deceive more effectively (e. g., Miller, deTurck, & Kalbfleisch, 1983; Siegman & Reynolds, 1983). This dissertation will explore several possible cognitive explanations for the impact of self-monitoring on deception. Particular attention will be given to potential differences in the planning procedure engaged in by high and low self-monitors as they prepare to deceive and engage in actual deception. LITERATURE REVIEW Self-Monitoringand Deception When confronted with situations in which it is necessary to deceive, high self- monitors apparently have greater control over the cues they emit and thus display more behavioral consistency whether lying or telling the truth. High self-monitors appear to exhibit fewer postural shifis, leg movements, vocalized pauses, and hand-to-head contacts than low self-monitors when lying, suggesting greater control over these behavioral aspects during deception (Riggio & Friedman, 1983). Self-monitoring also correlates positively with the ability to control speech rate, such that it is relatively constant whether the individual is lying or telling the truth (Siegman & Reynolds, 1983). Inhibitory control extends to deceptions about positive as well as negative emotions. Friedman and Miller- Herringer (1991) report that high self-monitors are more likely to attempt to mask their. expressions of happiness at winning a game when others are present and are more successfirl in these attempts. Low self-monitors, by comparison, appear to make little attempt to conceal their pleasure. Perhaps as a result of this cue control, high self- monitors are less likely to be detected when deceiving than low self-monitors (deTurck & Miller, 1990). Message content as well as cue displays may be influenced by self-monitoring. Goldsmith (1987) concluded that high self-monitors were more innovative than low self- monitors, and thus better able to construct novel deceptive messages. Stiff and his colleagues (Stiff, Corman, Krizek, & Snider, 1994) report that the decay-impulse intervention model that characterizes response latency during deceptive encounters provides a particularly good fit to the responses of truthful low self-monitors and deceptive high self-monitors, while being much less representative of the response patterns of the remaining two groups. These authors note that, during deceptive encounters, high self-monitors appear as though they are being truthful and appear to be paying little attention to their self-presentation. They recover more rapidly than low self-monitors 2 when confronted with probes during deceptive interactions and adjust their self- presentation, including the content of their deceptive messages, accordingly. Evidence suggests that high self-monitors may be better at both deceiving others and detecting deception. High self-monitors appear to be more accurate at decoding the facial affect of others, providing an avenue for detecting facial cues indicative of skepticism and distrust and adjusting deceptive messages accordingly (Mufson & Nowicki, 1991). This superior ability to decode facial affect also renders high self-monitors better able to detect deceptive cues produced by others (Brandt, Miller, & Hocking, 1980). Cronshaw and Ellis (1991) report that this sensitivity extends beyond facial cues to a general awareness of social cues and an ability to respond accordingly. Finally, Lassiter, Stone, and Weigold (1987) find that high self-monitors are more likely to remember information about others. This tendency may increase the likelihood that high self-monitors will be conscious of ' factual inconsistencies in the messages delivered by deceivers and thus more likely to judge inconsistent deceptive messages correctly. As can be ascertained from the previous discussion and fiom applications of the self- monitoring construct to domains other than deception (e.g., Bearden, Shuptrine, & Teel, 1989; Sypher & Sypher, 1983; Turner, 1980), the focus on self-monitoring has been almost exclusively directed toward detecting differences in nonverbal behavior rather than differences in actual message content. For example, Turner (1980) credits his finding that high self-monitors are more humorous almost exclusively to their ability to utilize effective joke delivery rather than to generate humorous material. Despite this narrow focus, it is possible that differences between high and low self-monitors stem, at least in part, from differences in cognitive structures that might be expected to influence message construction as well as manipulation of nonverbal behaviors. Findings that focus on both the encoding and decoding superiority of high self-monitors (e. g., Hosch & Platz, 1984; Siegman & Reynolds, 1983) suggest that such a process may be at work. This limited approach to self-monitoring may stem as much from methodological as from theoretical difficulties. Several authors (e. g., Briggs, Cheek, & Buss, 1980; Dillard & Hunter, 1989; Hoyle & Lennox, 1991; Lennox & Wolfe, 1984) have criticized the psychometric properties of both Snyder’s (1974) original Self-Monitoring Scale and Snyder and Gangestad’s (1986) shortened version. Criticisms have ranged from lack of conceptual correspondence to radical multidimensionality, both at the first- and second- order factor level. Dillard and Hunter (1989) strongly recommend against the continued use of these scales and caution that prior research based on these instruments must be treated with caution. In response to methodological and conceptual concerns, Lennox and Wolfe (1984) have developed a Revised Self-Monitoring Scale composed of two primary factors (Ability to Modify Self-Presentation and Sensitivity to the Expressive Behaviors of Others). Both Dillard and Hunter (1989) and Shuptrine, Bearden, and Teel (1990) report that this scale demonstrates acceptable reliability and second-order unidimensionality. It remains to be seen whether research findings relying on Snyder’s original or shortened measure can be replicated using the Lennox and Wolfe scale. Planning in Deceptive Situations One theoretical framework capable of organizing differences between high and low self-monitors is work on planning (e. g., Berger & Jordan, 1992; Waldron, 1990). According to this approach, when faced with a goal situation, people develop or call upon plans that detail the actions necessary for the attainment of the goal. These plans may differ in their degree of complexity. That is, a more complex plan might contain a greater number of specific steps and incorporate various contingencies that would allow the planner to deal successfiilly with unanticipated circumstances. The complexity of the resultant plan is believed to be determined in part by the individual’s prior experience with attempts to attain a similar goal, but also by the individual’s goal desire and planning desire (Greene, O’Hair, Cody, & Yen, 1985). Goal desire is defined as the strength of an individual’s desire to attain the goal under consideration. Planning desire, on the other hand, reflects an individual’s overall level of commitment to the planning process itself. These desires combine such that complex plans are typically generated only when both goal desire and planning desire are high. Plans, then, are cognitive structures that specify the actions the planner must take to achieve one or more goals (Berger, 1988). Plans can vary in their level of abstraction and the extent to which they specify alternative paths of action that the planner can pursue. As such, a plan for achieving the goal of successful deception might specify the actions of appearing calm, maintaining eye contact, providing a plausible account, and being sufficiently detailed so as not to arouse suspicion. The plan might also incorporate actions designed to achieve additional goals such as appearing likable and not insulting the interaction partner during the deceptive encounter. One would anticipate that no two individuals would form an identical plan for achieving a given goal. As Berger, Karol, and Jordan (1989) note, plans can differ along a number of dimensions. Some people will posses highly complex plans for achieving a given goal while others will possess simple plans or may possess no plan at all. These variations in complexity may manifest themselves in one of two ways. First, individuals may differ in the number of goal-relevant actions contained in particular paths of their plans (path complexity). Some will have highly specific, multi-step paths while others will have only vague, undifferentiated paths. A second way people may differ in complexity is in the number of alternative paths they have for achieving a particular goal (alternative complexity). Some will have a large arsenal of actions and approaches while others may be restricted to only a few. In addition to complexity, planners can also vary in the speed with which they are able to retrieve the actions that are contained in their plans (retrievability). Some may have a wide array of action-altematives but be unable to access these actions quickly enough to facilitate smooth, fluent performances. flanflsed Explmtions of Self-Monitoring Effects The plan-based approach can be used to generate at least three possible explanations for observed differences between high and low self-monitors. These explanations, to be discussed in turn, focus on (a) differences in desires and/or past experiences, (b) pursuit of multiple goals and/or ability to integrate multiple goals, and (c) modification of plans during actual encounters. It is possible that all, some, or none of these explanations can account for performance differences between high and low self-monitors. Experiences; and Desires Past experience. In the current context wherein participants are instructed to pursue the goal of successful deception, differences in prior experiences and desires must be considered. It is possible that high self-monitors have a greater number of prior experiences with deception than low self-monitors. Support for this conjecture is limited and mixed. Dobbins and his colleagues (Dobbins, Farh, & Werbel, 1993) report that, in a naturalistic study of the tendency to inflate grade-point averages, there was no difference between the amount and extent of inflation performed by high and low self-monitors. The authors did find, however, that low self-monitors who inflated their grade-point averages did so more consistently across situations. In contrast, Snyder and Gangestad (1986) indicate that ability to tell a lie with a straight face is a characteristic that reliably discriminates between high and low self-monitors. If it existed, a pool of past experiences would provide high self-monitors with a larger repertoire of tested action-alternatives than would be possessed by low self-monitors. Reliance on such a tested pool would be likely to improve the performance of high self-monitors as compared to low self-monitors. In addition, Greene et al. (1985) note that the strength of the structural relations between plan components is a function of the frequency and recency with which the plan is enacted. Greater past experience with deception would thus result in a more readily retrievable deceptive plan, an eventuality that would be particularly significant when opportunity to prepare for the deceptive encounter was not provided. Goal desire. It is also possible that liars in these contexts vary in the degree to which they experience goal desire. High self-monitors, as compared to low self-monitors, may judge the goal of successfirl deception to be more important. In naturalistic situations wherein the high self-monitor is more highly concerned with self-presentation, the goal of successful deception/avoiding detection may be central to the degree to which high self- monitors can productively manage the impressions others have of them. Snyder (1987) notes that high self-monitors tend to become depressed when others see them as incapable, whereas low self-monitors become depressed when they act in ways that are inconsistent with their beliefs. A deceptive context thus may provide the perfect venue in which to differentiate between high and low self-monitors. Whereas high self-monitors may place high value on the goal of successful deception because it allows them to avoid the depression associated with goal failure, low self-monitors may not value this goal ' because it forces them into a situation wherein their beliefs and overt behavior are in conflict. However, in the current context where the task involves deceiving a stranger who the participant will probably never see again and the deception is “permitted” or “required” by the experimenter, such self-presentational and consistency needs may be minimized and differences in goal desire may not vary significantly across levels of self- monitoring. Indeed, the work of Shaffer, Ogden, and Wu (1987) suggests that the prospect for firture interaction (or lack thereof) effects the communicative behavior of high but not low self-monitors, suggesting that the low probability of subsequent interaction may alter the behavior of high self-monitors in an experimental scenario involving deception. However, DePaulo, Kirkendol, Tang, and O’Brien (1988) offer an intriguing alternative explanation. Specifically, these authors argue that people who are more motivated to get away with their lies are less successfiil at doing so when nonverbal cues are accessible to detectors. This so-called “motivational impairment effect” could imply that, if high self-monitors place higher value on the goal of successful deception, this increased motivation may adversely effect their nonverbal cue production and render them more likely to be detected when lying. Whether or not this motivational impairment works against the previously discussed superior cue control of high self-monitors remains to be seen. Planning desire. A third possible experience/desire-based account for differences in the messages and behaviors of high and low self-monitors is that high self-monitors have greater planning desire than low self-monitors. In other words, high self-monitors may be more dedicated to the process of planning and see planning as a more valuable pursuit than do low self-monitors. Such a possibility is not inconsistent with the high self- monitor’s preoccupation with self-presentation, and would imply that, if given time to prepare, high self-monitors would show greater attention to planning activities than would low self-monitors. However, this general orientation toward planning may not map directly on to planning in the specific context of deception. In deceptive encounters, low self-monitors may exceed high self-monitors in planning desire if their goal desire is elevated. Specifically, low self-monitors attempting to deceive may find that the plans they retrieve are relatively simple and lacking in detail when compared to those retrieved by high self-monitors. If so, low self-monitors would be motivated to engage in the planning process in an attempt to develop and add specificity to their existing plans (Berger & DiBattista, 1992). To the extent that this specificity results in a level of complexity with which low self-monitors are unaccustomed to dealing, their performances should suffer as a result, accounting for the superiority of high self-monitors under rehearsal conditions (Greene et al., 1985). Multiple Goals An alternative explanation is that, although high and low self-monitors do not differ in their desire to achieve the goal of successful deception, high self—monitors have additional goal desires that are operating during these interactions. The additional pursuits might fi'uitfiilly be categorized as meta-goals in the sense that high self-monitors pursue the over-arching goal of successful impression management in all interactions. This cross- situational goal pursuit can be likened to what Berger (1994) terms the meta-goal of social appropriateness. High self-monitors would be expected to be more concerned with minimizing conflict-based and face-threatening interactions than would low self-monitors. One might then expect the message behavior and nonverbal behaviors of high self- monitors to differ from those of low self-monitors because the plans of high self-monitors incorporate additional strategies to achieve these impression management goals. Berger and DiBattista (1992) also suggest that increased concern with self-presentation while attempting to achieve social goals will contribute to the development of plans that are characterized by a greater number of contingencies, such that the communicator has something to “fall back on” should the initial plan appear not to be working. On a related point, it may also be possible that, because high self-monitors commonly pursue the impression-management meta-goal, they may have meta-plans that include action-altematives designed to achieve the meta-goal. These plans might include such actions as detecting what the bearer expects and adapting one’s behaviors to those expectations. For this reason alone, one might expect that the plans of high self-monitors would reflect a greater degree of complexity than the plans of low self-monitors. While Berger et al. (1989) found evidence that communicators with complex plans are less fluent than those with relatively simple plans, pursuit of the meta-goal alone should not account for this difference. Due to repeated enactment of the meta-plan across situations, the plan should become a script (Berger, 1994) and the high self-monitor should not experience any reduced fluency during social interactions. flanami BMr Modification A third possible explanation for differences between high and low self-monitors is that high and low self-monitors differ in their ability to modify their plans during actual encounters. Specifically, because high self-monitors demonstrate greater attention to the cues emitted by their interaction partners and greater ability to modify their self- presentation accordingly, they may be more likely to notice when a plan is not successful 10 at achieving a desired goal and be capable of modifying their plan accordingly. This modification would involve making on-line verbal or nonverbal alterations in the path currently being pursued or switching to a different path during the interaction. The work of Littlepage, Tang, and Pineault (1986) suggests that cue modalities may be more important than message content in distinguishing planned versus spontaneous lies. Specifically, these authors report that planned lies were more difficult to detect than spontaneous lies when presented in either video-only or audio-only format, but not when presented in transcript form. These findings indicate that planning may have a more noticeable impact on production of nonverbal cues as opposed to refinement of consistent verbal content. Berger and DiBattista (1993) also report that individuals who were thwarted in their original plans tend to maintain plan structure but reiterate messages that are less detailed and reflect changes in vocal intensity and speech rate. While these results may reflect the nature of the plan that was thwarted (direction-giving), this tendency to modify nonverbal cues bears monitoring in the current investigation. In sum, high self-monitors may use planning differently than low self-monitors for all, some, or none of the aforementioned reasons. The intent of the studies that follow is to devise a method of differentiating and testing these various explanations. One productive avenue for making these comparisons is through the use of structured and unstructured preparation time. Evidence suggests that high self-monitors tend to prepare more for deceptive encounters than do low self-monitors (Elliot, 1979) and also benefit more when given time to rehearse their deceptive messages than do low self-monitors (Miller, deTurck, & Kalbfleisch, 1983). Research by deTurck and Miller (1990) also indicates that those who have time to prepare and rehearse their deceptive messages are less likely to be detected, even by trained observers. Distinguishingflan-Based Accounts If differences between high and low self-monitors are due solely to differential ability to modify existent plans during interactions (Explanation 3), then no differences in 11 performance should be observed between high and low self-monitors as a fimction of planning time or planning condition. Moreover, if this explanation accounts for all of the variance in performance, then high and low self-monitors should report no differences in the content of their plans or the effort that they expend during the planning process. However, observation of the actual interaction should reveal clear differences in the messages and behaviors of high and low self-monitors. In particular, since high self- monitors are expected to be more attentive to cues indicating plan failure, they should notice these cues and modify their plans accordingly whereas low self-monitors, failing to notice such cues, should continue the present course of action without modification. Greene et a1. (1985) further suggest that increased cognitive effort should be reflected in such nonverbal variables as pupil dilation, increased speaker-tum latencies, lower sound- silence rations, shorter message duration, and reduced eye contact. Because plan modifications and production of deceptive messages are expected to require more cognitive effort on the part of low self-monitors, these low monitors might further be expected to demonstrate an increase in the display of these nonverbal behaviors relative to high self-monitors. If performance is at least in part due to differences in either experiential/desire variables (Explanation l) or the pursuit of self-presentation meta-plans (Explanation 2), planning condition should interact with self-monitoring to produce differences in message production. For example, if high self-monitors have greater planning desire than low self- monitors, high self-monitors should engage in more planning when provided with the opportunity to do so. As a result, the messages produced by high self-monitors in the planning conditions may appear more polished, smooth, and less prone to errors that might trigger detection (e.g., presence of adaptors, increased response latency). Degree of planning desire, goal desire, and past experience with pursuing deceptive goals also can be assessed using self-report data. Specifically, if high self-monitors experience greater planning desire than do low self-monitors, they should express higher agreement with such 12 statements as, “I prefer to put a lot of thought into what I will do and say in a situation before I am in the situation.” Similarly if high and low self-monitors do not experience differential goal desire in this situation, they should display similar levels of agreement with such statements as, “It was important to me that I was successfirl in convincing the other person that I was telling the truth.” Lastly, if high self-monitors have more extensive past experience with pursuing deceptive goals than do low self-monitors, they should express greater agreement with such statements as, “It is often necessary to bend the truth in order to accomplish one’s goals.” In addition to differences in self-report data, differences in planning desire should be reflected in the ways that people make use of planning time. If high self-monitors have greater planning desire than low self-monitors, observation of activity during planning time should reflect these differences. Specifically, high self-monitors should, on average, ‘ devote more time to planning than low self-monitors. Observation of planning activity may reveal that, while high self-monitors appear to spend the time planning, low self- monitors may instead engage in activities not related to preparation and planning (e. g., reading the newspaper). In addition, if one were to control for various other factors accounting for differences in messages between high and low self-monitors, the resultant plans that are enacted during deceptive encounters might still be expected to differ in path complexity and alternative complexity as a function of planning desire. As a result of the greater amount of time they spend planning, the plans of high self-monitors should be more complex than those of low self-monitors, other things being equal. Finally, if the effects of self-monitoring reflect differences in the pursuit of the meta- goal of self-presentation during planning, this phenomenon might be reflected in several indices. One approach would be to ask participants about their meta-goals and meta- plans; however, these structures may not be accessible through the use of self-report data. Berger (1994) notes that individuals may possess implicit goals in the sense that they pursue these goals within interactions without being consciously aware that they are doing 13 so. Given that the meta-goal of self-presentation is pursued across situations, it is unlikely that the high self-monitor is consciously aware of pursuing this goal in every situation. In a sense, such pursuit may become routinized and mindless through repeated enactment. Thus if we were to ask a high self-monitor to report if they pay attention to such goals, their reports may be inaccurate given this pursuit occurs below conscious awareness. Despite the potential unreliability of self-reports in this domain and the reduced likelihood that high self-monitors would recount having paid attention to self-presentation issue during planning, it should be possible to assess differences in the pursuit of the meta- goal of self-presentation by examining actual interactions. Ifhigh self-monitors are consciously pursuing the meta-plan of self-presentation, attention to such self-presentation factors should be evident in the dual focus of their verbal messages. That is, their messages should be characterized both by attention to the deceptive content and by such self-presentational concerns as attending to the face needs of the person being deceived and providing explanations for actions that might be misconstrued. Low self-monitors, in contrast, should demonstrate more singularity of focus, concentrating solely on attaining the goal of successful deception. One potential indicator of this dual focus may be evident in longer message durations for high self-monitors. Presumably, more complex (and thus lengthier) messages are required when pursuing multiple rather than single goals. As noted previously, high and low self-monitors tend to vary in the extent to which they benefit from having time to rehearse deceptive messages. Miller et al. (1983) conclude that high self-monitors are less likely to be detected when deceiving than low self-monitors, particularly when both groups are given time to prepare their messages. Specifically, while high self-monitors are less likely to be detected if given time to prepare, low self-monitors are more likely to be detected, resulting in a significant rehearsal by self- monitoring interaction but no main effect for rehearsal time. Such findings appear to run contrary to theorizing by Greene (1984) which posits that everyone will improve performance if given time to prepare because preparation time will facilitate activation and 14 assembly of procedural records for successful deception. Greene also notes that difficult integration tasks should result in disorganization of the output system, increasing the probability that lower level cues will function incoherently or contradict one another, thus inviting greater detection accuracy. To the extent that providing low self-monitors with rehearsal time encourages them to activate unfamiliar procedural records (e.g., those associated with self-presentation and with integrating self-presentation with other goals), rehearsal time should produce more difficult integration tasks for them, potentially resulting in greater message disorganization if integration is not complete at the onset of the interaction. Interestingly, however, nonverbal cues often associated with deception (e.g., message duration, speech errors, word/phrase repetitions, pauses, eye shifts, and adaptors) do not appear to account for performance differences as a fiinction of rehearsal time in the Miller et a1. (1983) research. Low self-monitors were found to produce fewer of these deception cues after rehearsal time than when they were asked to deceive without rehearsal, despite the fact that they were more likely to be detected in the rehearsal condition. One possible explanation for these findings is that rehearsal time impacts the quality of the verbal message more than it influences the display of nonverbal cues. Such an explanation is appealing in the light of findings that rehearsal time does not differentially impact the anxiety level of high and low self-monitors (Nfiller et al., 1983; DePaulo, Davis, & Lanier, 1980). Anxiety level had originally been proffered as an explanation for differential performance because it was known to impact the expression of nonverbal cues. If it is differences in the planning of message content that account for performance differences, one concern is whether it is possible to reduce planning differences by providing low self-monitors with information that can aid them in planning successful deceptive messages. In particular, it may be possible to stimulate planning desire, or at least the construction of complex plans in low self-monitors by providing them with 15 relevant information and instructing them to use this information in the preparation process. Berger and DiBattista (1992) report that, when individuals seek more information prior to planning, they are likely to develop more elaborate plans. Furthermore, the authors suggest that, if an individual retrieves an already elaborate plan from memory, their motivation to seek or make use of information would be lower. Whether the presence of such information differentially impacts high and low self-monitors could be determined by noting any differences in use of the material during rehearsal time and asking individuals to recount how they used the information, if at all, in constructing a plan. Providing information during planning may thus minimize differences in the abilities of high and low self-monitors to deceive effectively by allowing low self-monitors to develop plans that approach the elaborateness of those spontaneously retrieved by high self-monitors. In summary, three potential explanations for the differences between high and low self-monitors have been posited in this review. The explanations are not mutually exclusive in the sense that one or more may be necessary to account for performance differences. Explanation 1 proposes that high and low self-monitors may differ in the extent to which they experience planning or goal desire or the extent to which they have past experience with the pursuit of deceptive goals. This explanation can be tested directly using self-report data. Explanation 2 proposes that high and low self-monitors differ in the extent to which they pursue multiple goals in interactions and do so effectively. This explanation can be tested most fruitfully by examining the verbal and nonverbal content of actual interactions but may also be reflected in self-report data or recall data regarding the planning process. Explanation 3 proposes that high and low self- monitors differ in the extent to which they are able to modify their plans during actual interactions. This explanation can be tested by examining actual interactions for indications of cognitive load, particularly in the face of potential goal failure (e. g, voiced skepticism). 16 Exploring A Potenpial Confound One variable with the potential to complicate and invalidate the results of this approach is objective self-awareness. First explicated by Duval and Wicklund (1972), the theory of self-awareness posits that human beings are perpetually in a state of either objective or subjective self-awareness and that these states are precipitated by environmental cues. Those in a state of objective self-awareness see the self as an object apart from the environment and have a heightened sense of themselves as actors. In contrast, those in a state of subjective self-awareness direct their attention away fi'om the self and are less concerned with how the self appears to others. As a result, those experiencing objective self-awareness attribute greater responsibility to themselves for both positive and negative outcomes. In past research (e. g., Baldwin & Holmes, 1987; Webb, Marsh, Schneidennan, & Davis; 1989) such devices as pursuit rotors and small ' mirrors have been shown to be effective in inducing states of subjective self-awareness, whereas large mirrors, tape recorders, the presence of an audience, and videotaping contribute to the experience of objective self-awareness. Interactions between self-monitoring and self-awareness have been found in prior research (Ickes, Layden, & Barnes, 1978; Snyder & Monson, 1975), although the findings have been contradictory. Whereas Ickes et al. found that high self-monitors were less responsive to self-awareness manipulations than low self-monitors, Snyder and Monson report the opposite result. Webb et a1. (1989) argue that the key distinctions between these studies are in the modes by which self-awareness was induced and the experimental context. Specifically, these authors argue that if self-awareness inducing materials are cues for people to view themselves as others see them (e.g., large mirrors), high self- monitors should find the focus more informative. However, if the self-awareness manipulations and experimental context encourage a focus on the private aspects of the self, low self—monitors should benefit. Webb et al. report that, while high self-monitors attributed greater responsibility to themselves under conditions of heightened objective l7 self-awareness, they attributed less responsibility to themselves than did low self-monitors when objective self-awareness was not induced. These authors also report that these findings were particularly strong when participants were attributing responsibility for negative events. One explanation offered for these findings is that, while high and low self-monitors may both be vulnerable to self-awareness manipulations, high self-monitors cannot deal with the information provided by an inner focus productively, whereas low self-monitors fail to benefit from information gleaned from a self-presentational focus. This conclusion is supported by the work of Eichenhofer and his colleagues (Eichenhofer, Gerstein, Valutis, & Jankowski, 1987) on the impact self-monitoring and arousal. These authors note that the behavior of high self-monitors was unaffected by experimental attempts to increase or decrease the level of arousal experienced by participants. Again, the nature cf the experimental manipulation was implicated as a potential explanation for the results. Specifically, the behavior of high self-monitors may only be responsive to changes in arousal level when anxiety bears directly on task performance and there are anticipated social sanctions associated with task failure. Notably, both of these features are characteristic of deceptive encounters. In the current context, both inducing participants to prepare a message and videotaping them while they are delivering this message may reduce self-focused attention and produce an enhanced state of objective self-awareness. Duval and Wicklund (1972) theorize that people frequently will be objectively self-aware when preparing to send messages, and that exposure to information can further induce objective self-awareness if the individual anticipates having to convey that information to others (as would be the case when preparing a deceptive message, particularly with the aid of information). According to Duval and Wicklund, this enhanced self-awareness leads to a focus on discrepancies between actual and ideal self, or in the case of message behavior, to a focus on the correctness and consistency of beliefs and behaviors. In the context of deceptive 18 message production, such awareness may bring to light an unpleasant and arousing discrepancy for liars. Specifically, those engaging in deception may become aware of the discrepancy between their ideal self (truth teller) with their current selfas they may appear to others (deceiver), a discrepancy that may be particularly troubling for low self- monitors. Presumably such a realization would produce negatively valenced arousal. Given the demonstrated association between arousal and nonverbal cue leakage during deceptive encounters (e. g., deTurck & Miller, 1985; Greene et al., 1985; Zuckerman & Driver, 1985), we might expect that those experiencing a higher degree of objective self- awareness would be more prone to exhibit cue leakage associated with deception and thereby run the risk of greater detection. Furthermore, low self-monitors may be particularly vulnerable to this increased arousal. In contrast, truth tellers should not experience such increased arousal because their ideal selves and actual selves will not be brought into conflict. In summary, objective self-awareness offers an alternative explanation for the finding that high self-monitors are superior deceivers than low self- monitors, an explanation that demands examination in the current context. The next two sections discuss studies conducted to assess the viability of the arguments advanced in this review. The first study focuses on Explanation l and the extent to which pre—existing conditions such as past experience with deception and desire to plan differentiate between high and low self-monitors. The second study focuses on Explanations 2 and 3 and whether performance differences between high and low self- monitors can be replicated and accounted for on the basis of the quality of plan developed or the ability to modify plans during actual encounters. It also allows for the assessment of the role of objective self-awareness in experimentally contrived deceptive encounters. Several forms of information will be collected from each participant in an attempt to assess differences in planning orientation, plan content, and plan modification between high and low self-monitors. STUDY 1 Mann Overview of Desigp The first study was conducted to explore the proposed links between self- monitoring and the three dimensions of Explanation 1 (past experience, goal desire, and planning desire). After completing a self-monitoring questionnaire, respondents were asked to imagine themselves in a hypothetical situation to be used in the second study. They then completed a series of questionnaires designed to measure past experience with deception, level of goal desire (desire for achieving both the goal of successful deception and the goal of impression management) and level of planning desire (both general desire to plan and desire to plan within that particular context). Participants Eighty-two students enrolled in an introductory communication course completed the questionnaire packet during regular class periods. Students received extra course credit in exchange for participation. Procedures Participants first completed the self-monitoring questionnaire. They were then asked to imagine that they had signed up for an out-of-class experiment worth two extra credit points. The participants read a copy of the instructions to be used in the rehearsal condition of the second study (see Appendix C). After considering the situation, they were asked to complete a series of questionnaires detailing their responses to the situation and past experience with deception. When all participants had completed the questionnaires, they were debriefed and dismissed. Mafifials and Measures Self-monitorig sca_le§. The measure of self-monitoring used in Study 1 was the Revised Self-Monitoring Scale (Lennox & Wolfe, 1984).1 A seven-point response format bounded by “Always True/Always False” was used for all items. The Lennox and Wolfe l9 20 scale is composed of two component sub-scales: a seven-item measure of Ability to Modify Self-Presentation (AMSP) and a six-item measure of Sensitivity to the Expressive Behaviors of Others (SEBO). Prior research using the scale (e. g., Anderson, 1991; Lennox & Wolfe, 1984; Shuptrine et al., 1990) reports correlations between the subscales of .22, a reliabilites in the range of .70 to .77 and test-retest correlations ranging fi'om .52 to .55. mestionnaire on prior deceptive experience. The remaining questionnaires on past experience, goal desire, and planning desire were developed for use in this research and are shown in Appendix A. For the questionnaire on prior deceptive experience, respondents were asked to indicate the frequency with which they had lied in a variety of contexts. The eight contexts ranged from those focusing primarily on benefits to others (e.g., to avoid hurting another person’s feelings) to those focusing primarily on benefits to self (e.g., to get something that I wanted). A seven-point response format bounded by “Always/Never” was used for all contexts. mestionnaire on gpal desire. Two forms of goal desire were assessed in the first study. Three items were used to assess each respondent’s desire to achieve the goal of successfiil impression management in the hypothetical situation and five items were used to assess desire to achieve the goal of successful deception in the hypothetical situation. All items used a five-point Likert response format. Questionnaire on planning desire. This questionnaire assessed both general and specific orientation toward the planning process. Four items were used to measure the general value each participant placed on planning. Additionally, four items assessed the extent to which the participant would plan for the hypothetical deceptive encounter. Again, all items used a five-point Likert response format. 21 KGLSU—IQ grality of Measures Self-monitoring scales. The seven-item measure of AMSP and the six-item measure of SEBO were subjected to confirmatory factor analysis using the proposed two-factor structure. For the purpose of the factor analysis, self-monitoring scores of respondents in Study 1 were combined with those of respondents in Study 2, resulting in a sample size of 280. Factor loadings ranged from .43 to .80. With the deletion of a single item fiom the Sensitivity scale (Item 6 of the original measure), measures of both dimensions were internally consistent and parallel. For the Study 1 subgroup, the sub-scales of the Lennox and Wolfe self-monitoring measure correlated .45 (p < .001) with respective coeflicient or reliabilities of.79 (M = 35.83, S_D = 5.51) and .73 M = 25.74, SD = 3.93). Situafion—based questionnairfi. The questionnaires on past experience with deception, planning desire, and goal desire were combined and subjected to confirmatory factor analysis, with each scale representing either a single hypothesized factor or, in the case of planning and goal desire, two factors. However, for impression management goal desire and successful deception goal desire, the analysis indicated that, after the elimination of three bad items (factor loadings below .30), the remaining questions tended to hang together, such that a single five-item measure of goal desire was produced M = 16.70, SQ = 3.35, or = .76). The remaining factors retained their hypothesized structure except for Item 1 from the general planning desire factor. This item loaded more strongly on the deception planning factor and, given its conceptual relevance to deceptive planning in the current research, was moved to that cluster. Following these adjustments, the scales were found to be internally consistent and parallel. However, scale reliabilities were low. Whereas coefficient 0t for the past experience scale was an acceptable .86 M = 34.59, SD = 8.70), both the measure of general planning desire M = 10.65, SD = 1.86, or. = .55) and the measure of intent to plan for the hypothetical deceptive encounter (_M_ = 16.95, SJ; = 3.17, a = .65) produced less satisfying results. 22 Last Experience and Self-Monitoring It was anticipated that high self-monitors would have greater past experience with deception. T-tests for independent groups scoring high and low on both dimensions of self-monitoring were conducted with reported past experience as the dependent variable. Results indicated that AMSP showed no relationship with past experience. However SEBO appeared to be related to past deceptive experience. Examination of a scatterplot revealed that the relationship was non-linear. Specifically, those who were either very high or very low in SEBO (Ms = 36.52 and 37.40, respectively) reported greater past experience with deception as compared to those with more moderate levels Ms = 32.52 and 31.25). Given the extremity of high scorers, a four-way split on SEBO proved superior to a three-way split in capturing this trend. A one-way analysis of variance using a contrast analysis that reflected this interpretation was significant (1(78) = 2.68, p < .01). Goal Desire and Self-Monitoring It was predicted that high self-monitors would possess higher levels of impression- management goal desire and deception goal desire but that differences might be minimized by characteristics of the hypothetical situation (e.g., interaction with a stranger, relatively little sanction for failure to attain the goal). Again, t-tests for independent groups were conducted, but neither AMSP nor SEBO showed any relationship to goal desire, indicating that neither dimension of self-monitoring had an effect on desire to achieve the goals of successful deception and impression management. Whether this finding reflects characteristics of the hypothetical situation or the nature of self-monitoring is open to question. Planning Desire apd Self-Monitoring It was proposed that, whereas high self-monitors would place greater value on the general planning process, low self-monitors would exceed high self-monitors in their desire to plan for the specific deceptive encounter. Results of the independent groups t- tests indicated that neither AMSP nor SEBO showed any relationship to general planning 23 orientation. However, examination of the scatterplot revealed a relatively discontinuous difference between those high and low in AMSP in their desire to plan for the specific deceptive encounter. This variable was thus dichotomized using a median split, producing a significant difference between high and low modifiers (1(80) = 3.44, p < .005). ' Consistent with predictions, low modifiers evidenced more situational planning desire than high modifiers Ms = 17.98 and 15.70, respectively). The scatterplot revealed that SEBO also related to task-specific planning desire in a complex, nonlinear fashion. Specifically, after tetrachotomizing, the relationship assumed a dampened cubic form, with those scoring very low or moderately high on the SEBO measure (Ms = 17.85 and 18.06, respectively) showing more desire to plan for the deceptive encounter than those scoring in either the low moderate or very high range (both Ms = 16.17, t(78) = 2.57, p < .05). To examine the potential interaction between AMSP and SEBO on situation-specific A planning desire, SEBO was collapsed and recoded to reflect a linear trend. Specifically, the two high SEBO cells and the two low SEBO cells were each combined, resulting in a dichotomous variable with one group representing those high in situational planning desire and the other group representing those low in desire Ms = 17.94 and 16.17, respectively). When AMSP and the recoded SEBO were subjected to ANOVA, the interaction term did not approach significance (13(1, 78) = .31, as). Discussion The relationship between self-monitoring and past experience with deception is more complicated than originally anticipated and dependent upon the dimension of self- monitoring that is considered. AMSP appears to be unrelated to past experience with deception. In contrast, SEBO is related to past experience, but not in the manner originally predicted. Whereas it was hypothesized that high self-monitors would have more extensive past experience with deception, results revealed that those who are either very high or very low in sensitivity report greater past experience with deception as compared to those with more moderate levels. This finding has several implications. 24 First, it appears that differential past experience with deception is incapable of accounting for the performance differences observed between high and low self-monitors in prior research. Given that these studies relied on median splits on the self-monitoring variable, such experience differences would be equalized in the two groups. Second, the finding that those very high and very low in sensitivity are equivalent in past experience with deception does not explain why these two groups score high on this variable or necessitate that the same mechanism is driving both groups to engage in deception (or report engaging in deception) with greater frequency. However, given the current data, the nature of the mechanism remains unknown. The prediction that high self-monitors would possess higher levels of impression- management goal desire and deception goal desire was not borne out in the current research. Neither AMSP nor SEBO showed any relationship to the goal variables. It Was conjectured that differences, if they did exist, might be eliminated due to characteristics of the hypothetical situation (e.g., interaction with a stranger, relatively little sanction for failure to attain the goal). While this explanation remains plausible, it is an explanation that is equally applicable to prior research that has demonstrated performance differences between high and low self-monitors. Thus, while differences in goal desire may account for performance differences in naturalistic deceptive encounters between fiiends and intimates, such differences are incapable of accounting for the deceptive superiority of high self-monitors under less naturalistic laboratory conditions. These results firrther suggest that motivational impairments should not differentially affect high and low self- monitors in the current context. The data were not consistent with the hypothesis that high self-monitors would place greater value on the general planning process than would low self-monitors. Those high and low in AMSP and SEBO failed to differ in their general orientation toward planning. However, the hypothesis that low self-monitors would place more value on preparing for the specific deceptive encounter was at least partially borne out in the data. 25 Those scoring low on AMSP evidenced more situational planning desire than those scoring high. Furthermore, those scoring very low or moderately high on SEBO showed more desire to plan for the deceptive encounter than those scoring in either the moderately low or very high range. The two dimensions did not interact to affect desire to plan for deception In summary, those variables that predate an individual’s entrance into a deceptive encounter appear largely incapable of explaining observed differences in the deceptive success of high and low self-monitors. Past experience with deception and orientation toward the goal of successful deception appear largely unrelated to self-monitoring, at least in the way that it has traditionally been operationalized. Similarly, general dedication to planning appears to be equivalent in the two groups. Only desire to plan for the deceptive encounter was found to be related to self-monitoring status. This finding could account for the differences reported in prior studies if desire to plan resulted in the development of complex plans that could not be carried off effectively in actual interactions. This finding also fiirthers the possibility that the provision of goal-relevant information during planning may benefit low self-monitors, given that such a provision might reduce differences in ability to use planning time effectively. To assess these possibilities, actual deceptive interactions need to be examined for indications of planning behavior and the impact of this behavior on subsequent deceptive performance. These issues are the focus of Study 2. STUDY 2 Mcflnfl Overview of Desigp A mixed design was used in the second study. Message type (truthfirl vs. deceptive) was partially crossed with rehearsal condition (no rehearsal, rehearsal without information, or rehearsal with information). Specifically, all combinations were tested except the tl'llthfill, rehearsal with information condition. This condition was deleted from the design because it was impossible to obtain information that was equally or even remotely relevant to participants who were asked to tell the truth about a past experience. In addition, to check for the potential influence of objective self-awareness, a control group completed the two personality measures but did not participate in deceptive interactions. Participants The sample was composed of 198 undergraduates recruited from introductory and upper level communication classes. Participants signed up during class time for a convenient out-of-class session and received extra course credit in exchange for their participation. One hundred and sixty of these subjects (80 pairs) were assigned to the experimental group, while 38 students were assigned to the control condition. Two of the experimental group participants were later eliminated because they knew one another and had prior awareness of each other’s Spring Break activities. Of those remaining in the experimental group, 57 were male (36%) and 101 were female (64%). Procedures Three participants were recruited for each experimental interview session to increase the likelihood of having a sufficient number of participants for each session. The first two participants to arrive were escorted into a waiting area and asked to complete research consent forms. If a third participant anived, that individual was assigned to the control condition. She was taken to a different room where she completed a research consent form, the self-monitoring scale, and the measure of objective self-awareness. Upon 26 27 completion of this task, she was debriefed and dismissed. When participants assigned to the experimental condition had completed their consent forms, the experimenter collected the forrrrs and read a general description of the study. The experimenter answered any questions and then had participants determine who would be the interviewer and who would be the interviewee by drawing roles from a hat. Through this process, 36 males and 43 females were assigned the role of interviewer (46% and 54%, respectively) and 21 males and 58 females were assigned the role of interviewee (27% and 73%, respectively). After roles were determined, the experimenter distributed a self-monitoring questionnaire and biography sheet to each participant. When both participants had completed the forms, the experimenter returned to the room and escorted the interviewer to the interview room wherein microphones and video equipment were set up to record the interaction. The experimenter provided the interviewer with a list of sample questiOns (see Appendix B). This intervention was conducted to minimize the difference in interviewer preparedness as a result of the rehearsal condition into which the interviewee was to be assigned. The experimenter then told the interviewer that he or she would be interviewing the other student for three minutes about last Spring Break . A complete record of these and all other instructions delivered to experimental participants is provided in Appendix C. After confirming that the interviewer understood the procedure, the experimenter returned to the waiting area and delivered instructions to the interviewee. Depending on the condition to which the interviewee had been randomly assigned, the experimenter informed the interviewee that he or she would be either lying or telling the truth about his or her activities during the preceding Spring Break. All participants assigned to the deception condition were asked to try to convince the other student that they had take a Carnival Cruise to St. Thomas. The experimenter stressed the importance of convincing the other student and explained the nature of the three minute interview. If the interviewee was in the “no rehearsal” condition, he or she was then immediately escorted to the interview room. If the interviewee was in either the Rehearsal Without 28 Information or Rehearsal With Information condition, he or she also received instructions about rehearsal time. Both groups were told that they would receive three nrinutes to prepare and that they could use the time any way they liked. Participants who were assigned to the Rehearsal With Information condition were also provided with a packet of information about the cruise and about St. Thomas and were told they could use the information if they so desired. The experimenter then left the room and observed and timed the interviewee’s rehearsal activity through an adjacent one-way mirror. After three minutes had passed, the experimenter returned to the waiting room and escorted the interviewee to the interview room. When the interviewee was seated, the experimenter reiterated the interview instructions, started the video equipment, and left the room. The experimenter returned to the interview room after three nrinutes. At this time, the interviewer was provided with the Exit Questionnaire for Interviewers and asked to complete it in the interview room. The interviewee was escorted back to the waiting area and asked to complete the appropriate exit questionnaire for interviewees. As the interviewer and interviewee completed their exit questionnaires, the experimenter collected the materials and debriefed the participants. After the experimenter answered any questions, the participants were thanked and dismissed. Magdals anMd Measures Self-monitoring scale. As in the first study, the measure of self-monitoring was the Revised Self-Monitoring Scale (Lennox & Wolfe, 1984). Both the seven-item measure of Ability to Modify Self-Presentation (AMSP) and the six-item measure of Sensitivity to the Expressive Behaviors of Others (SEBO) were included. Items were scored on a seven- point format bounded by “Always True” and “Always False” Order and presentation of items was the same for both the experimental and control groups. Biography sheet. Each experimental participant completed a biography sheet. The biography sheet asked the students to indicate their names, majors, number of years at the 29 university, home towns, hobbies, and how they spent their Spring Break. This information allowed general level verification that the interviewees in the deception condition lied, and that those in the truthful condition told the truth (or at least remained consistent with their self-reports). In no instance had an individual assigned to the deception condition actually spent the preceding Spring Break on a Carnival Cruise to St. Thomas. Code sheet of rehearsal time activity. As noted above, interviewees in the rehearsal conditions were observed through a one-way mirror during the three minutes they were given to prepare. The experimenter conducting the observation wrote down what the interviewee was doing during each of the six 30-second intervals contained in the preparation segment. This observation provided data on observable on- and off-task behavior as well as use of the information packet when it was provided. Exit qu;stionnaire for interviewers. The exit questionnaire for interviewers was ' divided into four segments. The complete measure, and all other measures developed expressly for Study 2 are contained in Appendix D. The first section of the exit questionnaire was composed of three 7-point semantic differentials asking the interviewer to assess the veracity of the interviewee. The second segment asked the respondent to make judgments about the interviewee’s verbal behavior (six items), the third requested judgments about nonverbal behavior (six items), while the fourth asked for judgments about the interviewee’s overall behavior during the interview (five items). All items used a seven-point semantic differential response format. Exit qpestionpaires for interviewees. There were three forms of the exit questionnaire for interviewees, one for each of the three rehearsal conditions. The basic form was be completed by participants in the No Rehearsal condition. It contained five items asking respondents if they felt they were successfirl in convincing the interviewer that they were being truthfirl, the amount of anxiety they felt during the interview, and whether they feel their performance would have been improved if they had been given time to rehearse. Participants in the Rehearsal Without Information condition completed the 30 same basic questionnaire (minus the hypothetical rehearsal item) and were additionally asked to indicate their opinion about the impact of rehearsal time and the appropriateness of the amount of time given. This group also completed an open-ended response wherein they were asked to recall and write down all the things they thought about duringthe time they had to plan. Participants in the Rehearsal With Information condition completed the same questionnaire as those in the Rehearsal Without Information condition, with the additional requirement of being asked to evaluate the impact of the information on their performance. They also completed an additional open-ended response that asked them to recall their specific thoughts about the information provided and how they used it to plan, if at all. Objective self-flareness. Objective self-awareness was measured using the scale developed by Duval and Wicklund (1973). The scale contains ten hypothetical situations. The respondent is asked to assess the percentage of responsibility for the hypothetical outcome that should be ascribed to the respondent and the percentage that should be ascribed to the other participant in the scenario. The instructions stipulate that these percentages must sum to 100%. The assumption is that those experiencing heightened objective self-awareness will assume a greater degree of responsibility for the hypothetical events. Although Duval and Wicklund report both negative and positive hypothetical outcomes, only negative outcomes were used in the present research, given the finding of Webb et al. (1989) that high self-monitors show greater responsivity to objective self- awareness manipulations when considering negative events. The original 10 items were reproduced exactly, except that Item 4 of the original scale was replaced by the following item: “You ask your fiiend whether or not you should take an economics class that she had last term. Trying to impress you, she lies and says the class was super easy and she got a 4.0, when in fact she barely got a 1.5. You take her advice, enroll in the class, and get a 0.0. Because of this, you are unable to graduate this term.” This item was substituted because its content dealt with a hypothetical deceptive 31 encounter and because Item 4 on the original Duval and Wicklund measure had failed to produce results in the predicted direction during prior experimental manipulations (e. g., Duval & Wicklund, 1973). CodiMng of verbal and nonverbal cues. Because of technical difficulties and equipment malfirnctions, audio and video data were only obtained on 70 of the 79 interviews. Video-only data were obtained on two additional interviews, leaving seven interviews without video or audio records. On the basis of prior research, three verbal and nonverbal cues were selected to be coded from the successfirlly completed videotaped interactions. These cues and their definitions are shown in Table l and are based on those used by Miller et al. (1983). Table 1 Definitions of Verbal and Nonverbal Behaviors Message Duration The amount of time the communicator/deceiver spends talking, divided by the total length of the interaction. Response Latency The amount of time between the end of the interviewer’s question and the beginning of the communicator/deceiver’s answer, averaged over all interaction units in the interview. Adaptors Duration of time the communicator/deceiver uses his/her hands to fondle or manipulate objects and/or parts of his/her body, including clothing, divided by the total length of the interaction. These cues were selected because prior research has shown a link between these cues and actual deception and because these cues also have been shown to be related to the amount of cognitive effort required to produce the message and to the provision of preparation time. For example, both Cody, Marston, and Foster (1984) and Greene et al. (1985) report that the greater cognitive effort required in producing deceptive messages 32 results in longer response latencies for liars as opposed to truth-tellers. Greene and Lindsey (1989) indicate that the opportunity to prepare a message prior to its delivery results in significantly shorter response latencies. Furthermore, Miller et al. (1983) report shorter message durations as a result of rehearsal time and longer adaptors for spontaneous low self-monitors as compared to rehearsed high self-monitors. Finally, reviews such as that by Zuckerrnan, DePaulo, & Rosenthal (1981) indicate that these cues are some of the few that reliably distinguish between liars and truth-tellers across several studies. Res—uh; Quality of Measures Self-monitoring scales. As indicated previously, self-monitoring data from Study] and Study 2 were combined for the purpose of factor analysis, yielding a sample size of 280. This confirmatory analysis supported the two-factor solution and, with the deletion of Item 6 from the SEBO scale, produced two internally consistent and parallel scales. For the Study 2 subgroup, the sub-scales of the Lennox and Wolfe self-monitoring measure correlated .42 (p < .001). Coefficient or reliabilities were .79 and .81 for interviewer and interviewee AMSP, (Ms = 35.14 and 34.91, S_Ds = 5.33 and 5.62, respectively) and .71 and .69 for interviewer and interviewee SEBO Ms = 25.03 and 24.67, SDs = 3.79 and 3.67, respectively). Unless otherwise noted, all analyses involving self-monitoring reflect median splits on both dimensions. Exit questionnaire for interviewers. Confirmatory factor analyses were conducted on the four subscales contained in the exit questionnaire for interviewers, with each subscale hypothesized as separate factor. However, the initial analysis revealed that the three cue subscales were highly correlated (rs > .93) suggesting that a single underlying factor might be at work. This possibility was examined by collapsing the three cue scales into a single factor and maintaining the veracity judgment scale as a second factor. Five 33 items were successively eliminated from the composite cue scale to produce an internally consistent structure that demonstrated parallelism with the veracity judgment factor. This procedure yielded or reliabilities of .97 for the veracity judgment factor M = 10.03, SQ = 5.66) and .94 for the cue assessment factor (M = 36.73, _S_D = 15.42). MEective self-awareness. For this measure, the combined data from the experimental and control groups were used to conduct the confirmatory factor analysis. Given that prior research gave no indication of multidimensionality, the 10 items were hypothesized to form single factor and were subjected to confirmatory factor analysis accordingly. Initial analyses revealed a predominance of low factor loadings and a number of negative correlations among the items. Items were deleted successively in an attempt to eliminate these negative correlations. This procedure resulted in a four-item measure equivalent to the first factor extracted using an exploratory principle-axis factor analysis with varimax rotation. Three other factors with eigenvalues greater than one were also extracted, two of which contained only a single item. This first factor was retained as the measure of objective self-awareness, although its use is cautionary given an or reliability of only .45. These four items were internally consistent and demonstrated parallelism with respect to the veracity judgment factor. Codirpg of verbziand nonverbal cues. The videotapes of the interviews were coded by two independent coders who were blind to the hypotheses and the participants’ rehearsal conditions. However, because all deceivers produced the same lie (Spring Break in St. Thomas) it was impossible to train coders who were blind to the deception manipulation. Both coders coded eight of the 72 interviews (approximately 10%) to assess interrater reliability for all three cues. The intraclass correlations were 1.00 for message duration, .97 for response latency, and .92 for adaptor duration. After discussing discrepancies, the remaining interviews were divided between the two coders. Codingof MW. Both the code sheets from the rehearsal time observations and the interviewees’ self-reported use of rehearsal time and planning 34 materials were coded by the same coders who coded the verbal and nonverbal cues. Coders assessed the number of 30-second segments in which the interviewee was not engaged in off-task behavior (e. g., studying, doing a crossword puzzle) and the number of segments in which those interviewees in the Rehearsal With Information condition actively looked at the information packet. Self-report responses were coded dichotomously based on whether or not the interviewee indicated that s/he had actively planned during the time available and whether or not the interviewee indicated making use of the information provided when constructing the plan. Coders again coded eight interviews in common. Interrater reliabilities were 1.00 for both observed planning and observed information use. For the self-report measures, interrater reliabilities were also 1.00 for planning and for information use. Given that there were no discrepancies to discuss, coders divided the remaining questionnaires and observation logs and coded them independently. Replication of Previous Findingg Overview. The expressed goal of the current research was to test various explanations for findings obtained by previous researchers. Most notably, explanations were offered for the apparent superiority of high self-monitors as deceivers and the tendency of high self-monitors to reap greater benefits from rehearsal time than low self- monitors. Before these explanations can be evaluated, it is necessary to demonstrate that the findings of other researchers were replicated in the current study. Results relevant to these replicatory efforts are reported in the sections that follow. Given its absence from prior research, data from the rehearsal with information condition is omitted fi'om the current analyses where appropriate to increase comparability. Self-Monitoring and Detection Accuragy. Research by deTurck and Miller (1990) indicates that high self-monitors are less likely to be detected when lying than low self- monitors. To assess this hypothesis in the current study, a 2 (high/low interviewee AMSP) x 2 (high low interviewee SEBO) ANOVA was conducted on participants in the deception conditions. Results revealed no differences between high and low self-monitors 35 in the likelihood that they would be detected when lying. A lack of main effects suggested that those high in AMSP were no more likely that those low in ability to escape detection (E(l, 44) = 2.78, M). Similarly, those high in SEBO were equally likely to have their lies detected as were those low in sensitivity (F(1, 44) = 2.22, p_s_). However, the two dimensions of self-monitoring did interact with respect to detection accuracy such that those high in AMSP but low in SEBO were particularly likely to escape detection when lying (E(1, 44) = 4.83, p < .05). Brandt et al. (1980) report that high self-monitors are significantly more accurate in detecting deception than low self-monitors. In the current research, the 2 (high/low interviewer AMSP) x 2 (high/low interviewer SEBO) ANOVA on the deception sub- sample produced a nonsignificant main effect for AMSP(F(1, 44) = .38, as) on accuracy. SEBO did, however, discriminate between the two groups, yielding a significant main ' effect (£(1, 44) = 4.52, p < .05). Specifically, those high in sensitivity were more accurate than those low in sensitivity in judging deceivers. The two dimensions of self-monitoring did not interact to affect accuracy (E(1, 44) = .97, g), nor did cell means indicate that any non-standard interactions were present. Means and standard deviations for all groups are given in Table 2. For the sake of comparison, the impact of self-monitoring on detection accuracy was also explored for truthfirl interactions using the same 2 x 2 ANOVA format on the truthful subgroup. For interviewees delivering truthful messages, neither modification ability (F(1, 27) = .84, as) nor sensitivity (EU, 27) = .39, n_s) improved success, and no interaction emerged (E (1, 27) = 1.12, E). For detectors, results indicated a marginally significant main effect for modification ability, such that interviewers high in AMSP showed a tendency to be more accurate in judging a truth-teller’s veracity (E (1, 27) = 3.61, p < .07). Accuracy of truth judgments was not effected by the interviewers’ SEBO (EU, 27) = .00, n_s), nor was an interaction between the two dimensions observed (EU, 27) = 1.73, as). 36 Table 2 Impact of Self-Monitorirg on Detection Accurapy Interviewees High Ability Low Ability High Sensitive Low Sensitive Deception .40 (.50) .57 (.51) .54 (.51) .41 (.50) Truth .75 (.45) .87 (.35) .83 (.38) .77 (.44) Interviewers High Ability Low Ability High Sensitive Low Sensitive Deception .48 (.51) .48 (.51) .64 (.49) .35 (.49) Truth .94 (.25) .67 (.48) .83 (.38) .77 (.44) The Role of Rehearsal. Miller et al. (1983) report an interaction between rehearsal and self-monitoring, such that high self-monitors are less likely to be detected when provided with rehearsal time whereas low self-monitors are more likely to be detected when provided with preparation time. In an attempt to replicate this finding, a 2 (high/low modification) X 2 (high/low sensitivity) X 2 (no rehearsal/rehearsal without information) ANOVA was conducted on the deceptive subgroup using interviewer accuracy as the dependent variable. Results indicated no significant main effects or interactions for these variables. In addition, the data were inconsistent with the effect-coded contrast model reflecting the Miller et a]. interpretation both for AMSP (1(27) = 1.66, n_s) and for SEBO (1(27)= .55, n_s). Prior research indicates that rehearsal has no differential effect on the arousal levels of high and low self-monitors (DePaulo, Davis, & Lanier, 1980). In the current research, the effect of rehearsal was examined using 2 (high/low AMSP) x 2 (high/low SEBO) AN OVAs on the rehearsal subgroup with self-reports of the effect of preparation time on 37 arousal, degree of arousal experienced at the outset of the interview, and reports of changes in arousal during the course of the interview serving as dependent variables. Findings were consistent with prior research. Those high and low in modification ability did not differ in their reports of the effect of planning time on arousal (EU, 41) = .19, M). They also failed to differ in initial arousal (E(1, 41) = .10, 1‘_l§) or in change in arousal during the interview (_F_(1, 41) = 2.14, pa). Similarly, those high and low in sensitivity failed to differ in plan-based arousal (E(l, 41) = .30, as), initial arousal (EU, 41) = .04, g) and arousal change (E(1, 41) = 2.57, Q). There were no significant interactions between the dimensions of self-monitoring for any of the outcome variables considered. Overall, 82% of participants reported that preparation time either reduced or had no effect on their arousal level, whereas 78% reported either no change or reduction in their level of arousal during the interview. On average, participants reported moderate levels of arousal at the outset of the interview (M = 4.58 on the seven-point scale). Means and standard deviations for all groups are shown in Table 3. Table 3 IQpact of Self-Monitoringon Arousal in Rehearsal Conditions High Ability Low Ability High Sensitive Low Sensitive Planning Arousal 1.87 (.63) 1.73 (.83) 1.88 (.68) 1.71 (.78) Initial Arousal 4.48 (1.65) 4.68 (1.67) 4.50 (1.75) 4.67 (1.56) Arousal Change 1.87 (.87) 1.64 (.73) 1.63 (.82) 1.90 (.77) Greene et al. (1985) report that rehearsal reduces the likelihood that lies will be detected but has no effect on the detection accuracy for truthful statements. This finding was examined by using t-tests for independent groups to compare accuracy ratings for interviewees who either did not rehearse or rehearsed without information. Separate 38 comparisons were made for truthful and deceptive interviewees. Results replicated Greene’s findings for truthful interviewees, but not for deceptive ones. In truthfirl interactions, rehearsers (M = .79, SD = .43) and non-rehearsers M = .82, SD = .39) were equally likely to be believed (1(29) = .26, Q). There was also no difference in ability to elude detection for deceptive rehearsers M = .50, _S_D = .52) as opposed to deceptive non- rehearsers M = .29, SD = .47), with observed mean differences in the opposite direction of that found by Greene (1(29) = -1.16, pa). Prior investigations also suggest that high self-monitors prepare more for deception than low self-monitors (Elliot, 1979). This possibility was examined using both self-report and observational data. Using 2 x 2 ANOVAs, high and low self-monitors in the planning conditions were compared on the extent to which they believed that they would have done better if not given time to prepare and on their judgments of the effect of planning on ‘ ability to convince the other student. They also were compared on the extent to which they reported engaging in planning during the allotted time, and their opinions as to the adequacy of the planning time allotted. Additionally, observations of on-task behaviors during planning time were considered. Taking truthfirl and deceptive encounters together, the ANOVAs produced no significant main effects for AMSP. Interviewees high in modification ability did not differ in their judgments of the effect of planning time on ability to convince the other student (E(l, 41) = .03, ps). They also failed to differ on their judgments of the adequacy of the amount of preparation time provided (E(1, 41) = .77, as). Modification ability was not differentially related to either reported planning (E (1, 41) = 2.67, n_s), observed planning (13(1, 41) = .40, g), or predicted improvement in the absence of rehearsal (13(1, 41) = 1.14, M). Similar findings were obtained for the sensitivity dimension of self-monitoring. A lack of significant main effects indicated that those high and low in sensitivity gave similar ratings to the impact of planning time (F (1, 41) = .72, n_s), the adequacy of planning time (E(l, 41) = 1.44, pp), and the hypothetical result of lack of preparation (E(l, 41) = .72, 39 as). Sensitivity also failed to be related differentially to either self-reported or observed planning (Es(1, 41) = 2.22 and .14, respectively, n_s). Interestingly, however, the two dimensions of self-monitoring did produce a significant interaction for self-reported planning (EU, 41) = 10.02, p < .005) such that individuals who were either high on both dimensions or low on both dimensions were more likely to report engaging in planning during the time provided. A similar trend emerged for the judged impact of planning time, although this interaction was only marginally significant (E(l, 41) = 3.59, p < .07). Means indicated that those who were high on both dimensions or low on both dimensions were somewhat more likely to believe that preparation had improved their chances of convincing the other student. No other interactions approached significance. Means and standard deviations for both self-monitoring dimensions are shown in Table 4. Table 4 Impact of Self-Monitoring on Interviewee Preparation Low Ability High Ability Low Sensitive High Sensitive Low Sensitive High Sensitive Planning Effect 2.57 (.51) 2.38 (.74) 2.14 (.69) 2.69 (.60) Amt. of Time 1.93 (.47) 2.38 (.74) 2.00 (.58) 2.00 (.63) No Plan Effect 1.07 (.27) 1.38 (.74) 1.43 (.79) 1.44 (.63) Rep. Planning 2.00 (.00) 1.63 (.52) 1.86 (.38) 2.00 (.00) Obs. Planning 543 (1.16) 4.75 (2.34) 4.14 (2.27) 5.25 (1.29) Overall, 93% of interviewees indicated that having time to prepare had either not affected their chances of success or had improved their chances. Sixty-four percent felt that the three minutes allotted for preparation provided them with just the right amount of time to prepare. Seventy-six percent reported that they would not have done a better job of convincing the other student if not given rehearsal time. In terms of actual planning 40 activity, 91% of interviewees reported engaging in planning, and 67% were engaged in on-task behavior during all six of the 30-second planning time observation intervals. As noted above, these analyses reflect the combined reports of liars and truth-tellers. However, Elliot’s (1979) predictions relate only to deceivers. Thus, analyses were done to compare truth-tellers and deceivers with respect to the variables discussed above. A significant main effect for truthfirl versus deceptive condition was obtained for estimated preparation effect (H1, 37) = 4.40, p < .05), satisfaction with amount of time provided (E(l, 37) = 6.71, p < .05) and observed planning (E(1,37) = 11.05, p < .005). There was also a significant main effect for self-reported planning (F (1,3 7) = 6.47, p < .05). Examination of means revealed that deceivers felt they benefited more from planning, tended to endorse the same amount or an increase in preparation time, and were more likely to both report and be observed engaging in planning. The only significant two- or three-way interactions to emerge were the previously discussed two-way interaction between the self-monitoring dimensions for self-reported planning (E (1, 37) = 8.39, p < .01) and a three-way interaction between the dimensions and self-reported planning (E(1,3 7) = 4.57, p < .05). The nature of this interaction was such that truth-tellers who were low in modification ability and high in sensitivity reported the least planning, with truth-telling high modifiers/low sensitives and deceiving low modifiers/high sensitives reporting moderate levels. A one-way ANOVA with a contrast analysis reflecting this interpretation was significant (t(3 7) = 4.70, p < .001). The findings of this study are thus more complex than those reported by Elliot, both for deceivers and for truth tellers. Means and standard deviations for truthful and deceptive subgroups are given in Table 5. Cue Production. Prior research by several authors (e. g., Riggio & Friedman, 1983; Siegman & Reynolds, 1983; Friedman and Miller-Herringer, 1991) indicates that high self- monitors display more behavioral consistency when lying and telling the truth than do low self-monitors. Although this comparison is ideally conducted within a repeated-measures Table 5 41 Impact of Self—Monitoring, Truth-Telling, and Deception on Interviewee Preparation Planning Effect Amt. of Time No Plan Effect Rep. Planning Obs. Planning Planning Effect Amt. of Time No Plan Effect Rep. Planning Obs. Planning Truthfill Interviewees Low Ability High Ability Low Sensitive High Sensitive Low Sensitive High Sensitive 2.50 (.58) 2.33 (.58) 1.75 (.50) 2.33 (.58) 2.25 (.50) 2.67 (.58) 2.25 (.50) 2.33 (.58) 1.00 (.00) 1.33 (.58) 1.75 (.96) 1.33 (.58) 2.00 (.00) 1.33 (.58) 1.75 (.50) 2.00 (.00) 4.25 (1.71) 4.00 (3.46) 2.75 (2.06) 5.00 (1.00) Deceptive Interviewees Low Ability High Ability Low Sensitive High Sensitive Low Sensitive High Sensitive , 2.60 (.52) 2.40 (.89) 2.67 (.58) 2.77 (.60) 1.80 (.42) 2.20 (.84) 1.67 (.58) 1.92 (.64) 1.10 (.32) 1.40 (.89) 1.00 (.00) 1.46 (.66) 2.00 (.00) 1.80 (.45) 2.00 (.00) 2.00 (.00) 5.90 (.32) 5.20 (1.79) 6.00 (.00) 5.31 (1.38) design, the nature of the current data only allows for a comparison between independent groups of high and low self-monitors using t-tests. Behaviors coded from the videotaped interviews were used in this analysis in conjunction with the cue assessments completed by the interviewers. With regard to behavioral consistency, those low in modification ability did not significantly differ in the length of time they engaged in adaptors (1(3 5) = .90, g), the length of the messages they produced (1(34) = -.36, n_s), or the time it took them to respond to the interviewer’s questions (1(34) = .26, as) when lying versus telling the truth. However, they were rated by interviewers as producing significantly more cues to deception when lying as opposed to telling the truth (1(39) = -2.70, p < .05). 42 Those high in modification ability also failed to differ in adaptors (1(33) = .10, n_s), message duration (1(32) = -1.54, g) and response latency (1(32) = 1.69, as) when lying and telling the truth. In this case, interviewers also rated their nonverbal and verbal cue production similarly in both conditions (t(36) = -.37, pa). At least for modification ability, behavioral consistency generally characterizes both high and low modifiers. In the case of interviewer cue ratings, however, high modifiers did display more cross-situational consistency than low modifiers. The data on the sensitivity dimension of self-monitoring produced a similar pattern. Those low in sensitivity produced similar adaptor rates (1(3 6) = 1.18, pa), message durations (1(35) = -.72), and response latencies (t(35) = 1.14, g) when lying and telling the truth. Unlike low modifiers, however, interviewers also rated their cue displays similarly in both conditions (t(42) = -1.24, n_s). Interviewees scoring high in sensitivity‘did not differ in their adaptor rates (1(32) = -.38, n_s), the length of the messages they produced (1(31) = -1.32, M), or their delay in responding to interviewer questions (1(31) = .87, M). High sensitives also failed to receive different interviewer cue evaluations when lying and telling the truth (1(31) = -l .50, g). In the current research, cross-situational consistency when lying and telling the truth generally appears to characterize those scoring both high and low on the dimensions of self-monitoring. Only interviewer cue ratings for low modifiers demonstrated significant variability across truthful and deceptive contexts. Means and standard deviations for high and low self-monitors in truthful and deceptive interviews are presented in Table 6. The data on cue production (both observed and interviewer-evaluated) were also used in an attempt to replicate the finding of Miller et a1. (1983) that, during deceptive encounters, high self-monitors display fewer adaptors, shorter response latencies, and longer message durations than low self-monitors. Three 2 (high/low modifiers) x 2 (high/low sensitives) x 2 (no rehearsal/rehearsal without information) ANOVAs were conducted on the deception subgroup. Again, the data were not consistent with prior 43 research (see Table 7). Nonsignificant main effects indicated that those high and low in modification ability did not differ in the length of time they engaged in adaptors (E(1, 20) = .84, pa), the duration of their messages (F (1, 20) = .05, pg), or the length of time it took them to begin responding to interviewer questions (F(1,20) = .94, p_s_). Those high and low in sensitivity also produced similar adaptor lengths (EU, 20) = .27, as), message durations (Ij(l, 20) = .20, pa), and response latencies (E(1, 20) = .09, as). Table 6 Cue Consistency for High and Low Self-Monitors When Lyingand Tellingthe Truth Lying High Ability Low Ability High Sensitive Low Sensitive Adaptors 13.45 (19.54) 9.59 (12.34) 11.30 (15.24) 11.76 (17.72) Duration 54.86 (15.63) 51.68 (13.88) 54.27 (15.69) 52.14 (13.83) Latency 75.76 (24.42) 78.00 (34.10) 77.55 (33.73) 76.24 (24.99) Ratings 38.24 (15.51) 40.35 (17.14) 39.50 (16.05) 38.95 (16.69) Telling the Truth High Ability Low Ability High Sensitive Low Sensitive Adaptors 14.07 (15.02) 16.15 (24.39) 9.60 (10.56) 21.31 (25.52) Duration 47.33 (10.20) 49.92 (13.34) 48.87 (9.16) 48.50 (14.36) Latency 104.53 (62.50) 81.92 (52.76) 88.07 (39.17) 102.50 (77.37) Ratings 36.44 (14.47) 28.13 (10.76) 32.72 (12.54) 32.00 (14.77) Table 7 Cue Production in High and Low Self-Monitors During Deception High Ability Low Ability High Sensitive Low Sensitive Adaptors 15.64 (24.17) 10.86 (13.77) 12.40 (18.17) 14.23 (21.56) Duration 53.71 (16.53) 52.29 (13.67) 54.40 (16.57) 51.38 (13.20) Latency 69.43 (18.58) 81.00 (39.47) 75.27 (40.12) 75.15 (16.17) 44 In addition to the main effects discussed above, Miller et a1. (1983) also report that self-monitoring interacts with rehearsal, such that low self-monitors telling spontaneous lies produce more adaptors than high self-monitors delivering prepared lies. The preceding ANOVA yielded no significant main effect for rehearsal condition (EU, 20) = .43, M). However, significant two-way interactions emerged between rehearsal condition and modification ability (E(1, 20) = 8.98, p < .01) and rehearsal condition and sensitivity (F (1,20) = 10.58, p < .005). These were qualified by a significant three-way interaction between rehearsal condition, modification ability, and sensitivity (E (1, 20) = 8.55, p < .01) for adaptor duration. Examination of cell means indicated that low modifiers and high sensitives telling prepared lies produced significantly fewer adaptors than the remaining groups. This effect was qualified by the three-way interaction such that the greatest adaptor duration was observed for high modifiers/low sensitives telling prepared lies, ' while the shortest adaptor durations occurred for low modifiers/high sensitives delivering prepared lies and high modifiers/low sensitives producing spontaneous lies (_t_(20) = 4.85, p < .001). However, given the small cell sizes associated with this effect (particularly for the high adaptor cell), the instability of cell means makes strong interpretations unwise. Cell means and standard deviations are given in Table 8. Miller et a1. (1983) also note a tendency for rehearsal to shorten deceiver message duration but to have no effect on response latency. In the current research, independent groups t-tests indicated that rehearsal also had no effect on response latency (t(26) = -.49, n_s), with interviewees provided with rehearsal time M = 78.31, S_D = 41.44) taking as long to begin their responses as those who communicated spontaneously M = 72.53, SD = 18.58). However, rehearsal also failed to impact message duration (1(26) = -.12, as). Interviewees given rehearsal time M = 53.38, S_D = 13.87) produced messages comparable in length to those produced by spontaneous communicators M = 52.67, S_D = 16.22). 45 Table 8 Impact of Rehearsal and Self-Momtomrgpn Length of Adaptor Performance High Sensitivity Low Sensitivity High Ability Low Ability High Ability Low Ability No Rehearsal 20.20 22.00 2.00 16.20 (24.97) (26.87) (3.46) (15.12) Rehearsal Without 7.20 1.67 76.00 5.50 Information (10.21) (2.89) (0.00) (4.12) Tests of Planning Explanations Overview. Despite the fact that the configuration of results to be explained using plan-based accounts is somewhat different that originally anticipated, the explanations . developed in the preceding review should still be valid. In short, successful deceivers, when compared with their less successful counterparts, should demonstrate greater past experience with deception, greater goal desire, greater general planning desire, and less desire to plan for the specific deceptive encounter. Furthermore, successful deceivers should demonstrate greater plan complexity such that they demonstrate shorter response latencies to unexpected interview questions and deliver lengthier messages. However, provision of rehearsal materials may minimize differences between successfill and unsuccessful deceivers, having little effect on the prior group but improving the performance of the latter group. To accomplish impression management goals, successfirl deceivers should also better control the leakage of deceptive cues such as adaptors and receive more positive behavioral cue evaluations from interviewers. Finally, successful deceivers should be more skilled at plan modification during interaction, thereby demonstrating less variability in response latency over the course of the interview. In the current research, it has already been established that those high in AMSP but low in SEBO were most successful in 46 escaping detection when lying. It is this group, then, that should conform to the hypotheses outlined above. The remaining analyses focus on ascertaining whether or not this is the case. first Eaperience. Goal Desirwd Planmg. The data from Study 1 were re- examined to determine if the anticipated interactions between the two dimensions of self- monitoring emerged for past experience with deception, goal desire, and desire to plan in general and for the specific deceptive situation. One-way ANOVAs reflecting the predicted plan-based contrasts revealed that only desire to plan for the deceptive encounter conformed to expectations, with those high in modification ability and low in sensitivity expressing less desire to plan (t(78) = 2.52, p < .05). These most able deceivers did not differ from other groups in their past experience with deception, in their desire to convince the other student and maintain a good impression, or in their general dedicatiOn to the planning process. Engagement in planning activities and beliefs about planning were also assessed using data from Study 2. Separate ANOVAs were conducted for the five relevant dependent variables: (a) the interviewees beliefs about the hypothetical impact of lack of planning time, (b) beliefs about the effect of planning on success, (c) ratings of the adequacy of planning time, (d) extent of self-reported planning, and (e) extent of observed planning. Independent variables were AMSP, SEBO, and type of rehearsal cell (truth-no information, deception-no information, or deception-information). No main effects or interactions emerged for the hypothetical impact of lack of planning time. For impact of planning, ratings of planning time adequacy, and observed planning, the only significant main effect was for rehearsal cell. For the impact of planning (F (2, 33) = 4.24, p < .05), there was a linear increase in ratings of planning effectiveness from the truthful planning condition to the deceptive planning condition with information. For planning time adequacy (F(2, 33) = 3.94, p < .05), interviewees reported an increasing desire for more extensive planning time across those same conditions. Similarly, for observed planning 47 (E(2, 33) = 5.45, p < .01) amount of on-task behavior increased from the truthful condition through the deception condition with information. For self-reported planning, the rehearsal condition main effect was marginally significant (F(2, 33) = 3.23, p < .06) and reflected the trend outlined above. A significant AMSP x SEBO interaction also emerged for this variable (If (1, 33) = 6.87, p < .05) such that interviewees who were either high on both dimensions or low on both dimensions reported engaging in more planning. Again, the data were largely inconsistent with predictions. Only for self-reported planning was there a significant interaction between the self-monitoring dimensions, and the pattern of means was not consistent with particularly low situational planning desire in the high modifier/low sensitive subgroup. Indeed contrast analyses reflecting the hypothesized reduced situational dedication to planning for this subgroup were not significant for any of the dependent variables considered. Table 9 contains means and standard deviations for all comparisons. Cues Associated With Multiple Gpal Pursuit. Data from Study 2 were also used to assess whether interviewees high in modification ability and low in sensitivity would conform to expectations regarding cue production associated with plan complexity and impression management. Specifically, it was expected that this group would display shorter response latencies, longer message duration, and fewer adaptors. It was also predicted that they would receive more positive cue evaluations from interviewers. One- way ANOVAs with appropriate contrasts conducted on the deceptive subgroup indicated that the data were not consistent with the hypothesized model for response latencies (1(39) = -.70, pa), message duration (1(39) = -.64, as), adaptors (1(40) = -1.00, as), or cue ratings (1(44) = 1.55, as). As can be seen in Table 10, means were even opposite the predicted direction for response latency, adaptors, and message duration. High modifier/low sensitives showed a tendency to take longer to begin responding to interviewer questions, to display lengthier adaptors, and to deliver shorter messages. 48 Table 9 Planning Evaluations for Self-Monitoripg Subgroups Truth-Rehearsal Without Information Low Ability High Ability Low Sensitive High Sensitive Low Sensitive High Sensitive Nervous Plan 1.50 (.58) 2.00 (.00) 2.25 (.50) 1.67 (.58) Improve No Plan 1.00 (.00) 1.33 (.58) 1.75 (.96) 1.33 (.58) Plan Effect 2.50 (.58) 2.33 (.58) 1.75 (.50) 2.33 (.58) Plan Amt. 2.25 (.50) 2.67 (.58) 2.25 (.50) 2.33 (.58) Report Plan 2.00 (.00) 1.33 (.58) 1.75 (.50) 2.00 (.00) Observe Plan 4.25 (1.71) 4.00 (3.46) 2.75 (2.06) 5.00 (1.00) Deception-Rehearsal Without Information Low Ability High Ability Low Sensitive High Sensitive Low Sensitive High Sensitive Nervous Plan 2.00 (1.15) 2.67 (.58) 2.00 (.00) 1.67 (.82) Improve No Plan 1.00 (.00) 1.67 (1 . 15) 1.00 (.00) 1.50 (.84) Plan Effect 2.25 (.50) 2.00 (1.00) 3.00 (.00) 2.67 (.82) Plan Amt. 1.50 (.58) 2.67 (.58) 2.00 (.00) 2.00 (.63) Report Plan 2.00 (.00) 1.67 (.58) 2.00 (.00) 2.00 (.00) Observe Plan 5.75 (.50) 4.67 (2.31) 6.00 (.00) 5.17 (1.33) Deception-Rehearsal With Information Low Ability High Ability Low Sensitive High Sensitive Low Sensitive High Sensitive Nervous Plan 1.33 (.82) 1.00 (.00) 1.50 (.71) 2.00 (.58) Improve No Plan 1.17 (.41) 1.00 (.00) 1.00 (.00) 1.43 (.53) Plan Effect 2.83 (.41) 3.00 (.00) 2.50 (.71) 2.86 (.38) Plan Amt. 2.00 (.00) 1.50 (.71) 1.50 (.71) 1.86 (.69) Report Plan 2.00 (.00) 2.00 (.00) 2.00 (.00) 2.00 (.00) Observe Plan 6.00 (.00) 6.00 (.00) 6.00 (.00) 5.43 (1.51) Info Plan 3.00 (.00) 3.00 (.00) 2.50 (.71) 2.86 (.38) Nervous Info 1.33 (.82) 1.00 (.00) 2.50 (.71) 2.14 (.69) Report Info 200 (.00) 2.00 (.00) 2.00 (.00) 2.00 (.00) Observe Info 5.83(.4l) 6.00 (.00) 5.00 (1.41) 5.14 (2.27) Table 10 Nonverbal Cue Performgnce of Self-Monitorig Subgroups Response Latency Message Duration Adaptors Cue Judgments 49 Low Ability Low Sensitive High Sensitive 71.87 91.14 (11.11) (15.70) 53.53 47.71 (13.93) (13.95) 9.33 10.14 (20.56) (52.79) 43.27 34.88 (17.22) (16.65) High Ability Low Sensitive High Sensitive 87.17 71.20 (29.12) (15.53) 48.67 57.33 (14.18) (15.94) 17.83 12.20 (33.36) (19.40) 29.71 41.56 (11.70) (15.81) The Role of Rehearsal and Information. An additional issue to be explored was whether provision of rehearsal time and, in particular, rehearsal with information, could diminish the superior deceptive ability of high modifiers/low sensitives. To examine this, one-way ANOVAs using Scheffe multiple range comparisons were conducted independently on unsuccessfirl and successful deceivers. Only two cue variables discriminated across rehearsal conditions. The first, adaptors, emerged only when deceivers were detected (F (2, 19) = 14.95, p < .001). Deceivers who were not provided with preparation time produced significantly more lengthy adaptors than did deceivers who either prepared without information or with information. Interviewer cue judgments also distinguished between rehearsal conditions, but only when deceivers escaped detection (F(2, 22) = 6.83, p < .005). Deceivers who rehearsed without information received significantly less favorable cue evaluations than either non-rehearsers or deceivers who rehearsed with information. Means and standard deviations for all cue comparisons are given in Table 11. 50 Table 11 Cue Production of Successful and Unsuccessful Deceivers Across Rehearsal Conditions Successful Deceivers No Rehearsal Rehearsal With Rehearsal Without Information Information Adaptors 5.20 (5.43) 19.50 (29.14) 5.00 (6.51) Message Duration 57.10 (16.42) 56.33 (9.00) 57.50 (16.53) Response Latency 67.20 (17.31) 72.00 (17.52) 79.50 (38.81 Cue Ratings 25.50 (7.03) 37.57 (10.28) 21.67 (8.52) Unsuccessful Deceivers No Rehearsal Rehearsal With Rehearsal Without Information Information Adaptors 36.00 (19.18) 3.14 (2.97) 10.60 (8.24) . Message Duration 43.80 (12.91) 50.86 (17.35) 51.11 (13.82) Response Latency 83.20 (17.89) 83.71 (55.73) 80.44 (19.23) Cue Ratings 57.20 (10.13) 52.71 (13.94) 48.18 (11.29) ANOVAs conducted on the cues and judgments yielded no significant main effects or interactions for the self-monitoring dimensions or rehearsal conditions on message duration, response latency, or interviewer cue ratings. Similarly, one-way ANOVAs indicated that the data were largely inconsistent with the effect-coded model predicting that provision of rehearsal time without information would benefit high self monitors but not low monitors and that provision of information would equalize improved performance for these two groups. For adaptors, the hypothesized model was also rejected. However, the straight ANOVA produced significant two-way interactions with rehearsal condition for both modification ability and sensitivity (5(2 ,31) = 5.82, p < .01 and E(2, 31) = 6.89, p < .005, respectively) as well as a significant three-way interaction (F(2,31) = 5.76, p < .01). 51 Examination of the means for these interactions indicated that adaptors were most frequent for low modifiers without rehearsal and high modifiers rehearsing without information. Adaptors were least frequent for low modifiers in either of the two rehearsal conditions (1(3 8) = 2.03, p < .05). For the sensitivity dimension, adaptors were least frequent among low sensitives rehearsing with information and high sensitives rehearsing without. They were most plentiful for low sensitives rehearsing without information and high sensitives who did not rehearse (1(3 8) = 2.20, p < .05). Again, cell sizes were too small to make valid interpretations of the three-way interaction. However, examination of the means in Table 12 suggests that this effect may be entirely attributable to the single high modifier/low sensitive outlier who rehearsed without information. Table 12 Adaptor Durations for Self-Monitors Across Rehearsal Conditions Low Ability High Ability Low Sensitive High Sensitive Low Sensitive High Sensitive No Rehearsal 16.20 22.00 2.00 20.20 Rehearsal No Info 5.50 1.67 76.00 7.20 Rehearsal Info 6.17 11.00 12.50 9.20 On-line Modification It was also conjectured that successful deceivers might be believed because they were more capable of making smooth on-line adjustments in their messages during deceptive encounters. One way that this superiority was expected to manifest itself was through more consistent response latencies across the length of the interaction for interviewees who were high in self-monitoring. To examine this hypothesis, within- subject variance in response latency was calculated for each interviewee. This variance 52 was then used as the dependent variable in ANOVA with self-monitoring dimensions and provision of rehearsal time as the independent variables. No significant main effects emerged, although the two-way interaction between the self-monitoring dimensions was significant (F(1,3 5) = 5.11, p < .05). Examination of the means for this interactiOn revealed that individuals high in one self-monitoring dimension but low in the other (Ms = 13491.43 and 8495.33) displayed greater inconsistency in their response latencies when deceiving than did those who were either high or low on both dimensions Ms = 3067.00 and 3801.00). A Look at Objective Self-Awareness Recall that objective self-awareness was offered as of potential confounding variable with the capability of explaining observed difference without reference to plan-based accounts. Specifically, it was noted that inducing careful preparation of a message and videotaping delivery of this message may have reduced self-focused attention and produced an enhanced state of objective self-awareness, particularly when information was provided during rehearsal time. The resulting state was then expected to lead to a focus on discrepancies between actual and ideal self, or in the case of message behavior, to a focus on the correctness and consistency of beliefs and behaviors. In the context of message production, such awareness might have brought to light unpleasant and arousing discrepancies for liars, particularly for those low in self-monitoring who are more likely to endorse belief-behavior consistency. The cues produced as a result of such arousal were then expected to render low self-monitors more likely to be detected when deceiving. Several strategies were used to examine the potential for objective self-awareness to serve as an alternative explanation for the findings. First, objective self-awareness scores for experimental participants (those who were videotaped) were compared to the scores of participants in the control group using independent groups t-tests. The analysis indicated no difference between the two groups (1(114) = 1.10, M). Indeed, the pattern of means suggested that the control group M = 57.28, S_D = 21.37) was somewhat higher in 53 objective self-awareness than the experimental group (M = 52.89, S_D = 19.63). Interestingly, however, those in the control group scored significantly higher on both the modification ability and sensitivity dimensions of the self-monitoring scale than did participants in the experimental group (t(114) = 2.66, p < .01 and t(114) = 2.05, p < .05, respectively). Those in the control group obtained average scores of 37.76 (512 = 5.27) and 26.13 (S_D = 3.57) on the two dimensions, whereas those in the experimental group obtained average scores of 34.86 (S_D = 5.64) and 24.65 (S_D_ = 3.69). Uncorrected correlations between objective self-awareness and the self-monitoring dimensions were not significantly different in experimental and control groups, although the correlations between objective self-awareness and SEBO approached significance (; = 1.58, p < .12). Given unreliability in measurement, particularly for the objective self-awareness scale, it is likely that the true correlations do differ significantly from one another (see Table 13). ' This result suggests that the experimental procedures may have altered the relationship between the sensitivity dimension of self-monitoring and objective self-awareness. Within the experimental context, the data revealed that those scoring high M = 54.76, SD = 18.71) and low (M = 51.10, S_D = 20.44) on the modification dimension of the self-monitoring scale did not differ significantly in objective self-awareness (t(77) = -.83, g). In addition, those scoring high (M = 53.95, S_D = 20.28) and low M = 51.80, _S_D_ = 18.76) on the sensitivity dimension also failed to differ in objective self-awareness (1(7 7) = -.48, n_s). Deception condition also failed to bring about differences in self- awareness (1(77) = -.77, n_s), with liars (M = 54.36, S_D_ = 20.71) evidencing as much objective self-awareness as truth-tellers M = 50.89, S_D = 17.65). As for cues possibly associated with the experience of objective self-awareness, only self-reported arousal appeared to be differentially associated with this variable. There was a marginally significant tendency for those low in objective self-awareness (M = 4.78, S_D = 1.46) to report experiencing higher levels of arousal at the outset of the interview (1(77) = 1.88, p < .07) than those high in objective self-awareness (M = 4.11, S_D = 1.74). 54 Table 13 Correlations Between Self-Monitoring Dimensions and Objective Self-Awareness Experimental Group Self-Awareness Modification Ability Sensitivity Self-Awareness 1 .00 . 10 .20 Modification Ability .06 1.00 .56 Sensitivity . 10 .41 1 .00 Control Group Self-Awareness Modification Ability Sensitivity Self-Awareness 1 .00 -. 13 -.40 Modification Ability -.08 1.00 .50 Sensitivity -.22 .3 7 1 .00 Note. Lower triangle = Uncorrected; Upper triangle = Corrected for Unreliability The predicted three-way interaction between the two self-monitoring dimensions and the cell of the experimental design (truth/deception status and rehearsal type) was examined using ANOVA. No significant main effects or interactions emerged. Means suggested that modification ability interacted with experimental condition to produce differences in objective self-awareness, but, as predicted, this effect occurred only in the condition where deceivers rehearsed with information. Specifically, low modifiers M = 39.03) deceiving after rehearsing with information tended to be less objectively self-aware than high modifiers M = 62.36) deceiving after rehearsing with information (t(15) = -1.91, p < .08). No other differences approached significance. Factors Contributipg to Successfirl Message Production Telling Lies. As previous results indicate, plan-based explanations appear to offer modest explanatory value for the deceptive superiority of high modifiers/low sensitives. As a result, it was judged desirable to see which if any of the variables included in this study were capable of distinguishing between successful and unsuccessfirl liars and 55 truth-tellers. To do so, two discriminant analyses were conducted, one for deceptive messages and the other for truthfirl ones. Accuracy of the interviewer’s judgments was used as the criterion variable. In total, 14 variables were entered into the analysis, and a stepwise algorithm maximizing Wilks’ A was used. For deceivers, seven variables met the criteria for entry into the equation. The discriminant fimction yielded an eigenvalue of 2.51 and a canonical correlation of .85. The x2 of 47.06 was significant (p < .001). Variables entered into the equation discriminating between successful and unsuccessful deceivers were cue ratings of interviewers, variance in response latency, rehearsal condition, interviewee gender, change in arousal during the interview, reported arousal level, and interviewee AMSP. This filnction correctly classified 91% of undetected liars and 95% of detected liars. The fimction coefficients, structure matrix, and changes in A associated with each included variable are given in Table 14. Successfirl deceivers received more positive cue ratings from interviewers, showed less variability in their response latencies, were more likely to indicate that their level of arousal decreased during the interview, and were more likely to be male. Liars who escaped detection were also more likely to be deceiving without rehearsal time, to indicate lower levels of arousal prior to the interview, and to score higher on the modification ability factor of the self-monitoring scale. However, only cue ratings met the .30 structure coefficient cut-off for discriminatory meaningfirlness, suggesting that interviewer perceptions of nonverbal and verbal cues account for most of the discrimination between successfill and unsuccessful liars. The associated function coefficient greater than 1.00 indicates some multicolinearity between the predictors. Indeed, when considered independently, both cue ratings (1(46) = -7.36, p < .001) and rehearsal condition (t(46) = -2.11, p < .05) differ significantly between successfirl and unsuccessfirl deceivers. Means and standard deviations for discriminating variables are included in Table 15. Note that it is the goal of deceivers to be judged incorrectly, whereas it is the goal of truth-tellers to be judged correctly. 56 Table 14 Discriminant Function Coefficients, Structure Matrix, and Changs in A Deceivers Function Structure Coefficients Coefficients Reduction in A Cue Ratings 1.10 .74 .58 Latency Variance .45 .11 .05 Rehearsal Condition .39 . 14 .02 Interviewee Gender .37 .20 .02 Arousal Change -.33 -.02 .02 Initial Arousal -.28 . 13 .02 Interviewee AMSP -.27 -. 12 .01 Truth-Tellers Function Structure Coefficients Coefficients Reduction in A ‘ Cue Ratings 1.05 .72 .50 Response Latency .70 .36 .10 Arousal Change -.38 .04 .03 Initial Arousal -.37 -.02 .03 Telling the Truth. For truth-tellers, the picture was somewhat more restrictive. While the same 14 variables were entered into the analysis, only four discriminated between successful and unsuccessfirl truth-tellers. The derived function yielded an eigenvalue of 1.94 and a canonical correlation of .81. The x2 of 24.81 was significant (p < .001), and the fimction correctly classified 100% of misjudged and 96% of correctly identified truth-tellers. Variables that discriminated between successfill and unsuccessful truth-tellers were interviewer cue ratings, response latency, reported level of arousal at the outset of the interview, and change in arousal level during the interview. Truthfirl interviewees who received more positive cue evaluations from interviewers, demonstrated shorter response latencies, indicated that their level of arousal decreased during the interview, and reported higher levels of arousal at the beginning of the interview were 57 more likely to convince the interviewer that they were indeed being truthful. Function coefficients for both cue ratings and response latency met the criterion for meaningfulness, and both significantly discriminated between successful and unsuccessful truth-tellers when examined independently (1(29) = 5.05, p < .001 for cue ratings and 1(25) = 2.52, p < .05) for response latency). Table 15 Variables Discriminating Between Successfirl and Unsuccessful Liars gprd Truth-Tellers Cue Ratings Response Latency Latency Variance Rehearsal Condition Interviewee Gender Arousal Change Initial Arousal Interviewee AMSP Deceivers Correct Incorrect 51.52 27.96 (11.95) (10.90) *** *** 7557.29 4078.59 (12796.31) (7055.35) 2.26 1.76 (.81) (.83) 1.83 1.60 (.39) (.50) 1.74 1.68 (.92) (.80) 5.22 4.80 (1.04) (1.53) 34.29 36.16 (5.56) (5.86) Diagram Replication of Prior Research Truth-Tellers Correct Incorrect 28.04 50.67 (10.14) (28.04) 83.74 156.25 (44.37) (95.72) *** *** *** *** *** *** 1.84 2.00 (.80) (.89) 3.68 3.33 (1 .75) (1.63) *** *** Summary of Finding; As noted previously, the primary goal of this research was to test a range of explanations for findings that had been obtained in previous studies. Unfortunately, the current research failed to replicate several of those findings. 58 Specifically, there was no difference between those scoring high and low on either dimension of self-monitoring in the likelihood that they would be detected when lying or telling the truth. Those high in AMSP were also equivalent to those low in ability in their success at detecting deception. However, SEBO discriminated between the two groups in the predicted direction. Consistent with prior research, those high in sensitivity were more accurate than those low in sensitivity in judging deceivers. There was also no difference in ability to elude detection for deceptive rehearsers as opposed to deceptive non-rehearsers and no evidence that high self-monitors were more likely to engage in planning than low self-monitors for either truthful or deceptive encounters. Similarly, there was only limited evidence of greater cross-situational behavioral consistency on the part of high self-monitors. While no differences emerged for either self-monitoring dimension on rates of cue production coded from the videotapes, those low in modification ability were rated by interviewers as producing significantly more cues to deception when lying as opposed to telling the truth, indicating greater variability for this group. Rather than finding that high self-monitors display fewer adaptors, shorter response latencies, and longer message durations than low self-monitors, the data indicated no difference in cue production for these groups. Provision of rehearsal also failed to impact message duration such that interviewees given rehearsal time did not produce shorter messages than spontaneous communicators. The prediction that low self-monitors delivering spontaneous lies would produce more adaptors than would high self-monitors delivering prepared lies also proved to be problematic. Although spontaneous liars scoring low on both dimensions produced more adaptors, so did high modifiers who were low in sensitivity and delivering prepared lies. On a positive note, the current data were consistent with prior research indicating that rehearsal has no differential effect on the arousal levels of high and low self-monitors for either dimension of the scale, and that 59 rehearsal has no impact on detection accuracy for truthful statements. The data also provided a replication for the finding that rehearsal has no effect on response latency. When other complex interactions were predicted, the data were again inconsistent. Whereas it was predicted that self-monitoring would interact with rehearsal, such that low self-monitors telling spontaneous lies would be more likely to be detected than high self- monitors presenting prepared lies, this effect did not emerge. Instead, a three-way interaction between rehearsal, modification ability, and sensitivity in the present project revealed that those high in modification ability and low in sensitivity were most successfirl at eluding detection regardless of whether they were given time to prepare. In contrast, those low in modification ability who were given time to prepare were most likely to be detected, regardless of their level of sensitivity. Thus, while it appears that low modifiers are harmed by rehearsal, rehearsal does not augment the deceptive superiority of high ' modifiers. The general failure to replicate was characterized by more than simple null findings or opposite effects. For example, there was a significant interaction with respect to detection accuracy such that those high in AMSP but low in SEBO were particularly likely to escape detection when lying. An additional interaction revealed that, for self-reported planning, people who were either high on both self-monitoring dimensions or low on both dimensions were more likely to report engaging in planning during the time provided. These same individuals showed a greater tendency to acknowledge the role preparation in improving their odds of convincing the other student. Two additional marginally significant interactions emerged in the planning domain. First, those scoring low on both dimensions were most likely to be observed planning during the observation segments, whereas those high in modification ability but low in sensitivity were observed planning least frequently. This finding is intriguing given the finding that the latter group was most likely to escape detection when lying. Perhaps high modifiers/low sensitives are aware of their skill at deception and thus do not feel the need 60 to plan as much as do their less skilled counterparts. The second interaction relates clearly to this explanation, with participants scoring low on both dimensions least likely to indicate that they would have been more successful if not given time to prepare. Explanations for Discrepancies. Before moving on to a discussion of analyses specific to the current investigation, it may be informative to explore some of the potential reasons for the inconsistencies with prior research. There are several possible explanations, most involving characteristics of the study design. First, the nature of the interaction between the interviewee and the interviewer are unique to the current investigation. In most prior research, the interviewer was either a confederate of the experimenter (e. g., Greene et al., 1985; Miller et al., 1983; Stiff et al., 1994) or the deceiver lied to a video camera without interacting with any interviewer (e.g., DePaulo et al., 1988; Ekman, 1988; Riggio & Friedman, 1983). As a result, variations in intervieWer characteristics (e.g., knowledge of topic lied about, suspicion, quality of questions asked, motivation to detect deception) introduced substantial variability into the current research, variability that was not present in prior investigations. Such variability likely elicited substantial variations in interviewee behavior within the cells of the experimental design, rendering it more difficult to identify significant trends in the data. This increased variability is consistent with research by Stiff and Miller (1986) indicating different cue production in liars confronted with suspicious versus accepting interviewers. Despite lack of correspondence with prior findings, however, it should be noted that the current design bears a stronger resemblance to actual deceptive interactions than the other studies discussed above. Another characteristic of the current investigation that sets it apart from prior research is the method by which accuracy judgments were obtained. In the current investigation, the naive interviewer participating in the face-to-face interaction was the only individual to provide a deception judgment on a particular interviewee. In other research (e. g, DePaulo et al., 1988; Ekman, 1988; Miller et al., 1983) detectors were 61 organized into small groups with each group viewing a videotape of several deceivers. Such an approach does not allow detectors to interrogate deceivers. This design characteristic is unfortunate considering prior findings that indicate differential nonverbal cue production in deceivers confronted with probing versus those who were not probed (e. g, Buller, Comstock, Aune, & Strzyzewski, 1989; Stiff & Miller, 1986). The findings of Buller and his colleagues (Buller et al., 1989; Buller, Strzyzewski, and Hunsacker, 1991) also bear on accuracy judgments. These authors found that probers were more prone to a truth bias than were detectors who were not allowed to probe regardless of whether sources were lying or telling the truth. They also note that observers of interactions are significantly more accurate at detecting deception than are conversational participants, presumably because the cognitive load of interpersonal interaction diverts attention from cue processing. Interestingly prior research largely devoid of probing generally reports detection accuracy approximating chance levels in deceptive encounters (c.f., Zuckerman et al., 1981). In the current research, interviewers were accurate 81% of the time when judging truth-tellers and 48% of the time when judging deceivers. A filrther characteristic of many prior studies that differentiates them from the current research is the opportunity for detectors to view (DePaulo et al., 1988; Miller et al., 1985) or interact with (Buller et al., 1989) multiple deceivers. Recall that that interviewers in the present investigation interviewed and rated only one interviewee. This procedure removed the chance for comparison between similar messages delivered by multiple sources. The difference is potentially significant given that prior studies have demonstrated that detection accuracy improves significantly when judges are provided with opportunities to view both deceptive and truthfirl messages delivered by the same communicator (Brandt et al., 1982; O’Sullivan, Ekman, & Friesen, 1988). Ifthis finding results from the impact of comparison as much as from the impact of familiarity, 62 participants in less naturalistic studies may be at an advantage in accurately judging the behavior of interviewees. An additional explanation for the lack of comparability stems from the nature of the lie required of interviewees. Most prior research required deceivers to offer brief deceptive responses to emotion-eliciting stimuli, whereas the present study required a lengthy lie about factual activities engaged in by the deceiver. Fer example, both Miller et a1. (1983) and Ekman (1988) required deceivers to conceal emotional reactions to such gruesome stimuli as medical training films and slides of disfigured burn victims. Related to this issue, Ekman (1988) notes that lies generally fail for one of two reasons. First, deceivers fail because their emotional reactions produce nonverbal cue leakage that betrays them. Lies about emotion-eliciting stimuli are particularly vulnerable to such cue production. Ekman filrther argues that the second reason deceivers fail is because they have not adequately prepared their lies and thus produce verbal messages that are self-. contradictory. Lies about factual information are particularly vulnerable to these sorts of inconsistencies. To the extent that the self-monitoring construct taps sensitivity to and ability to modify nonverbal behavior (scale items indicate this focus), this variable would be expected to differentiate between successfirl and unsuccessfirl deceivers and detectors in research involving emotional lies. However, the scale’s appropriateness and ability to differentiate factual lies like those delivered in the current investigation is far less clear cut. Measurement of self-monitoring may also account for discrepancies between the present and prior investigations. Dillard and Hunter (1989) note that, given the radical multidimensionality of the original Self-Monitoring Scale (Snyder, 1974) and its demonstrated lack of parallelism, research that has shown significant results using the scale should be reconsidered. Most prior research establishing the link between self- monitoring and deception has relied on either the original Snyder scale or on Snyder and Gangestad’s (1986) shortened version (e. g, deTurck & Miller, 1990; Miller et al., 1983; Riggio & Friedman, 1983; Siegman & Reynolds, 1983). In contrast, the current study 63 uses the Revised Self-Monitoring Measure (Lennox and Wolfe, 1984), a scale with superior psychometric properties. While Snyder and Gangestad note that their measure and the Lennox and Wolfe measure share approximately 50% of their reliable variance, it is possible that the aspects of self-monitoring accounting for differences in prior research are contained in a portion of the variance that is unique to the Snyder scale. To determine if the scale used could account for the differences in findings, the analysis attempting to replicate that rehearsal by self-monitoring interaction obtained by Miller et a1. (1983) was rerun using Snyder and Gangestad’s (1987) scale as the single factor measure of self-monitoring (the method used by Miller et al.).2 Miller et al. also label high self-monitors as those in the t0p 20% of the sample distribution and low self- monitors as those in the bottom 20%. AN OVAs were run using both a median split on self monitoring and the polarized scores used by Miller. Results indicated no main effects or interactions for self-monitoring on accuracy for either the median split or the extremity scores. However, examination of cell means did indicate non-standard interaction in the data for both groups. For the median split group, there was a tendency for low self- monitors to be detected when given preparation time, whereas the remaining groups did not differ from one another (t(27) = 2.04, p < .06). For the extreme group, it appeared that the pattern of means conformed to Miller’s findings. Specifically, high and low self monitors did not differ in ability to elude detection when rehearsal was not provided (Ms = .33 and .50, respectively). In addition, high self-monitors improved as a result of rehearsal M = .00) whereas low self-monitors suffered M = 1.00). However, given small cell sizes (a = 9), it was impossible to confirm this interpretation with statistical tests. Despite this, instrumentation decisions involving the self-monitoring variable still appear capable of accounting for some of the lack of correspondence between this research and its predecessors. A final variable thought to influence cue displays is motivation to lie. Ekman (1988) argues that, when motivation is high (as was supposedly the case in his research), liars 64 display shorter response latencies and less frequent adaptors. He argues, however, that when motivation is low, liars display longer response latencies, more frequent adapters, and shorter message duration (deTurck & Miller, 1985; Greene et al., 1985). However, Stiff et al. (1994) also find longer response latencies in deceivers despite using a situation wherein motivation to lie was high (avoiding being caught cheating). While the findings of Stiff and his associates pose problems for Ekman’s explanation, the explanation still has interesting implications for the current research. While motivation was not measured directly in Study 2, motivation to deceive in the corresponding hypothetical scenario was assessed in Study 1. The mean score on the factor measuring desire for goal success was 3.34 (S_D_ = .67) on a five-point scale, indicating that the majority of participants agreed that they would be somewhat concerned with deceiving successfirlly and would put effort into trying to convince the other student. While participants were not told, as they had been in the Ekman (1988) research, that their performance was indicative of future success in their chosen profession, they were offered an opportunity to double the number of extra credit points they received for participating if they were successful in convincing the other student. This manipulation probably elevated motivation to lie in the current research to moderate levels, less than that experienced in the research by Ekman and Stiff et a1. (1994), but more than that experienced in studies conducted by Miller, Greene, and their colleagues (deTurck & Miller, 1985; Greene et al., 1985; Miller et al., 1983). If highly motivated deceivers produce shorter response latencies when lying and less motivated deceivers produce longer responses, perhaps moderately motivated communicators show no difl‘erence in response latency when lying and telling the truth. Adequacy of Plan-Based Accounts Despite the fact that plan-based hypotheses were re-articulated to conform to the nature of successfirl deceivers in the current research, the data largely failed to conform to expectations. In terms of what interviewees brought with them to the current research 65 (past experience, goal desire, and specific/general planning desire) only desire to plan for the specific deceptive encounter achieved significance. Successful deceivers (high modifiers/low sensitives) indicated reduced desire to plan. This finding is consistent with planning explanations in that these deceivers may be aware of their skill and thus not be motivate to exert effort in planning. One explanation offered for this greater skill was that these deceivers would have greater past experience with deception and thus have more developed and solidified plans in memory (Berger, 1994). However, this explanation is not consistent with the null finding concerning past experience with deception. Given that past experience failed to vary across subgroups, high modifiers/ low sensitives must have achieved (or inherited) their superior deceptive ability in some other fashion (e. g., through the more studied observation of other deceivers). At least on the level of cue correlates, the data were also not consistent with explanations involving the pursuit of multiple goals. High modifiers/low sensitives failed to display shorter response latencies, longer message duration, and fewer adaptors. They also only showed a marginally tendency to be rated by interviewers as displaying fewer of the nonverbal and vocalic cues typically associated with deception. However, at least two factors may mitigate against the abandonment of the multiple goal explanation at this juncture. First, it is possible that the nonverbal cues examined in the current investigation are only tangentially related to the pursuit of multiple goals and thus incapable of discriminating successful versus unsuccessful deceivers. Content analyses of actual messages may provide a more thorough evaluation of this hypothesis. Second, it is possible that nonverbal cues are still indicative of multiple goal integration, but that the precise cues necessary were not evaluated in the current research. While this explanation runs contrary to cue expectations outlined in the planning literature (e. g., Berger et al., 1989; Greene et al., 1985), it remains a slight possibility. In a broader context, the rehearsal manipulation used in the present study also failed to elicit the cues anticipated on the basis of prior investigations. In short, the effect-coded 66 model based on the planning explanation was rejected for message duration, adaptors, and interviewer cue ratings. Only for response latency did marginal support for the model emerge, and this finding held only for the modification ability dimension of self- monitoring. For adaptors, the hypothesized model was also rejected. Although Obtained two- and three-way interactions indicate variability in adaptor production across conditions, this pattern appears largely unrelated to ability to elude detection. Other Findings Objective Self-Awareness. As was noted previously, objective self-awareness was proffered as a potential competing explanation for findings from previous research indicating the debilitating impact of rehearsal on low self-monitors. Not only is this variable apparently incapable of explaining the results, but obtained findings cast serious doubt on the theory’s soundness. First, although such activities as planning, delivering a message, and being videotaped are all supposed to elevate objective self-awareness, data from the current research suggested higher rates of objective self-awareness in the control rather than the experimental group. This result is intriguing given that control group participants were not exposed to any stimuli likely to heighten their objective self- awareness. Objective self-awareness also failed to discriminate on the basis of deception condition or level of self-monitoring. Only for self-reported arousal was there a marginal tendency for those low in objective self-awareness to report experiencing greater arousal prior to the interview. The joint consideration of self-monitoring status and planning condition further suggested that only the provision of information elicited differences in objective self-awareness, and this variability only occurred for high and low modifiers who were also high in sensitivity. Such a restricted finding can hardly be taken as an explanation for prior findings, especially given that the provision of information during rehearsal was unique to the current study. A final concern involving objective self-awareness involves the way it is assessed experimentally. Although Duval and Wicklund (1972) offer an impressive list of strategies 67 whereby objective or subjective self-awareness might be elicited, they are much less forthcoming in providing an assessment of the level of awareness being experienced. The situational evaluation questionnaire (Duval & Wicklund, 1973) proved woefirlly inadequate in the current research. To obtain an internally consistent factor structure, more than half of the scale items had to be discarded, and the remaining items produced an unacceptably low reliability coefficient. In short, the objective self-awareness explanation may still hold, but until better means are developed for assessing the extent to which awareness is experienced, findings involving the construct can only be treated as suggestive. Chapacteristic of Successful Liargrnd Truth-Tellers. Given that the plan-based explanations at least appeared incapable of explaining the breadth of the superiority of high modifiers/low sensitives, it was considered valuable to determine what factors . discriminated between successfirl versus unsuccessful liars and truth-tellers. Successfirl deceivers received more positive cue ratings from interviewers, demonstrated less response latency variability, were more likely to indicate that their level of arousal decreased during the interview, and were more likely to be male. Liars who escaped detection were also more likely to be deceiving without rehearsal time, to indicate lower levels of arousal prior to the interview, and to score higher on the modification ability factor of the self-monitoring scale. For truth-tellers, those who received more positive cue evaluations from interviewers, demonstrated shorter response latencies, indicated that their level of arousal decreased during the interview, and reported higher levels of arousal at the beginning of the interview were more likely to convince the interviewer that they were indeed being truthful. These findings have a couple of implications in the current context. First, apparently plan-based accounts, while they may account for some of the variance in success (particularly for deceivers), are incapable of entirely accounting for differences between successfirl and unsuccessfill communicators. Furthermore, the finding that arousal tended 68 to decrease over time for successfirl liars and truth-tellers suggests that global cue measurements averaged over an entire interaction may not capture the essence of the distinction between successfirl and unsuccessfill communicators. Sequential analyses or time-sampling methods may be better able to discriminate among these groups. It is also intriguing that, while the vast majority of interviewees indicated that provision of rehearsal time improved their likelihood of convincing the interviewer, interviewer ratings suggested the opposite effect, with spontaneous liars eluding detection more fi'equently. It may be that planning time would benefit deceivers if it were utilized properly, but deceivers appear largely incapable of using preparation time effectively despite believing that positive outcomes are associated with planning. CONCLUSION This dissertation represents a first effort at going beyond demonstrating the superiority of some types of people as deceivers and moves toward offering an explanation for why these effects are observed. Its status as a first effort no doubt accounts for the magnitude of many of the findings. However, this research has also revealed what can be considered limitations in prior research, both in terms of instrumentation and methodology. Apparently blanket statements regarding the deceptive superiority of a particular subgroup are uncalled for and should be qualified with reference to such factors as nature of the lie, motivation to deceive, and the context within which the lie was delivered and evaluated. Future research should consider the mechanisms behind such cross-situational variability. Furthermore, if plan-based explanations of deception are to be given a thorough test, consideration must be given to the content of both plans and messages delivered, ideally in a context that allows for more creativity and variability in the messages and approaches devised by deceivers. 69 APPENDICES Appendix A Measures Developed For Use In Study 1 (* Indicates item omitted from final measure) Past Experience With Deception In the past, I have lied to avoid hurting another person’s feelings. In the past, I have lied to avoid getting in trouble. In the past, I have lied to get something I wanted. In the past, I have lied to avoid doing something I didn’t want to do. In the past, I have lied to make someone feel better. In the past, I have lied to impress other people. In the past, I have lied to get other people to like me. . In the past, I have lied just to see if I could get away with it. General Planning Desire 1. I prefer to be spontaneous rather than put a lot of thought into what I am going to say— H 2. I like to think about different ways I might approach somebody before I actually initiate an interaction. 3. If I’m going to ask somebody to do a favor for me, I often rehearse the request in my mind before I make it. 4. I often plan different possible things to say in a situation so that I’ll have something to say no matter what happens. Deception Task Planning Desire 1. I would spend a lot of time thinking about what I was going to say to the other student. 2. I would spend the time reading information about St. Thomas if it were available. 3. I think I could lie just as effectively if I weren’t given time to prepare. (-) 4. If I put a lot of thought into what I was going to say, the other student would be more likely to know I was lying. (-) Goal Desire: Impression Management 1. * It would be important to me that the other student like me. 2. It would be important to me to control what the other student thought about me. 3. * I would be concerned that the other student would think less of me if s/he discovered that I was lying. Goal Desire: Successful Deception 1. * IfI didn’t lose the two extra points, it would not bother me if the other student knew I was lying. (-) 2. It would be important to me that the other student believe what I said. 3. I would want to be successful in convincing the other student even if it weren’t worth extra points. 4. I would put a lot of effort into trying to convince the other student. 5. It really wouldn’t matter to me whether I could successfirlly deceive the other student. (-) WflQ‘MPP’Nf" 70 Appendix B Sample Questions Provided for Interviewers General Questions PWNS“ “>19.“ 9. 10. ll. 12. How long were you there? What kinds of things did you do? Did you have fun? Would you like to do the same thing next year, or would you want to do something different? What was the weather like? Did you meet anyone new over break? How long did it take you to get there? How was it? What one word would best describe your Spring Break? Is this the kind of thing you typically do over break? Where did you go? How much money did you spend? F ollow-Up Questions 9:59.“? Could you tell me more about that? What else did you do? What else happened? Could you be more specific? Anything else? General Instructions These questions are just intended to get you started. Feel free to think up any other questions you might want to ask, either while you’re waiting or during the interview. 71 Appendix C Instructions to Participags in Study 2 Instructions to Interviewers In this part of the study you’ll be asking the other student about what he (she) did over Spring Break. He (She) will be instructed to either lie or tell the truth about where he (she) went and what he (she) did. If you can successfully detect if the person is lying or telling the truth, you will receive two additional extra credit points. To help you, I’ve given you a sheet of questions you might want to ask. Feel free to think up other questions while you’re waiting or during the interview. Do you have any questions? Instructions to Interviewees: Deception Condition In part two we are interested in your ability to convince another student that you spent your Spring Break in a different place and did different things than you actually did. Specifically, we would like you to try to convince the other student that you took a Carnival cruise to St. Thomas. We have already instructed the other student that you may be trying to deceive him (her), so he (she) may be unlikely to believe you. If you are . successful in convincing the other student that you are telling the truth, you will receive two additional extra credit points. If the other student correctly judges you, they will receive two points. To make it more challenging, we have given the other student some sample questions and instructed him (her) to think up other questions to ask you about where you went and what you did. You and the other student will be given three nrinutes to discuss your Spring Break activities. Any questions? Instructions to Interviewees: Truth Condition In part two we are interested in your ability to convince another student that you spent your Spring Break exactly as you said you did on the biography sheet. The catch is that we have already instructed the other student that you may be trying to deceive him (her), so he (she) may be unlikely to believe you. If you are successful in convincing the other student that you are telling the truth, you will receive two additional extra credit points. If the other student correctly judges you, they will receive two points. To make it more challenging, we have given the other student some sample questions and instructed him (her) to think up other questions to ask you about where you went and what you did. You and the other student will be given three minutes to discuss your Spring Break activities. Any questions? Instruction to Interviewees: Rehearsal Without Information Since it is important that you successfirlly convince the other student, I’ll now give you a few minutes to rehearse and prepare the message that you’re going to deliver. Feel free to use this time however you want. I’ll be back in a few minutes. 72 73 Appendix C: Instructions to Participants in Study 2 (cont’d) Interview Instructions: Both Parties As you know, you’ll now have three nrinutes to discuss your Spring Break. Your interaction will be videotaped for the purpose of the research. I’ll turn on the cameras now and you can begin talking as soon as I leave the room. Any questions? Instructions to Interviewees: Rehearsal With Information Since it is important that you successfully convince the other student, I’ll now give you a few nrinutes to rehearse and prepare the message that you’re going to deliver. Feel free to use this time however you want. It’s been our experience that it often helps people to rehearse if they have something to help them. As a result, we’ve provided you with a brochure about Carnival cruises that you can look over while preparing for the interview. I’ll be back in a few nrinutes. Debriefing Statement In this study we are interested in seeing how people’s awareness of the impressions others have of them and their ability to modify these impressions affects their ability to lie or tell the truth and to detect when other people are lying to them. We’re also interested in whether having time to rehearse a message affects performance and if different people. use this rehearsal time in different ways. We will be using the videotape we made of you to look at differences in people’s verbal and nonverbal behavior when lying or telling the truth. Do you have any questions or comments about the study? Appendix D Measures Developed for Use in StudLZ (* Indicates item was deleted from final measure) Exit Questionnaire for Interviewers: Veracity 1. How truthfirl were the interviewee’s responses to you questions? 2. How likely is it that the interviewee spent Spring Break as s/he said she did? 3. How confident are you that the interviewee was telling the truth? Exit Questionnaire for Interviewers: Overall Belgrvior 1. * Tense 1 2 3 4 5 6 7 Relaxed (-) 2. Consistent 1 2 3 4 5 6 7 Inconsistent 3 Secure 1 2 3 4 5 6 7 Insecure 4. Changing 1 2 3 4 5 6 7 Constant (-) 5 Sure I 2 3 4 5 6 7 Unsure 6 Anxious 1 2 3 4 5 6 7 Calm (-) Exit Questionnaire for Interviewers: Answers 1. Clear 1 2 3 4 5 6 7 Vague 2. Specific 1 2 3 4 5 6 7 General 3 . Plausible 1 2 3 4 5 6 7 Implausible 4. * Unbelievable 1 2 3 4 5 6 7 Believable (-) 5. Inconsistent 1 2 3 4 5 6 7 Consistent (-) 6. Direct 1 2 3 4 5 6 7 Evasive Exit Questionnaire for Interviewers: Body Movements 1. * Tense l 2 3 4 5 6 7 Relaxed (-) 2. * Consistent 1 2 3 4 5 6 7 Inconsistent 3 . Secure 1 2 3 4 5 6 7 Insecure 4. * Changing 1 2 3 4 5 6 7 Constant (-) 5 Sure I 2 3 4 5 6 7 Unsure Exit Questionnaire for Interviewees: No Rehearsal 1. Do you think the other student believed that you were telling the truth? 2. How aroused and anxious were you when you began your conversation with the other student? 3. How did your level of anxiousness change as the interview progressed? 4. Do you think the other student doubted that you were being truthfirl? 5. If you had been given time to prepare before the interview, how do you think it would have affected your performance? 74 75 Appendix D: Measures Developed for Use in Study 2 (cont’d) Exit Questionnaire for Interviewees: Rehearsal Without Information 1. 2. How did having time to prepare for the interview affect how nervous you were? Do you think you would have done a better job of convincing the other student if you hadn’t had time to prepare? Items 3-6 acre identical to items 1—4 of thgnrecedingguestiomaire How do you think that having time to prepare affected your chances of convincing the other student? When you think about the preparation time you were given, would you say there was too much time, just the right amount, or not enough time to prepare? During the 3 minutes you had to prepare, what did you do? What did you think about during this time? Exit Questionnaire for Interviewees: Reheaflrl Without Information 10. 12. Items 1-8 are identical to items 1-8 of the preceding questionnm How do you think having the cruise information affected your ability to prepare for the interview? How did having the information affect how nervous you were before the interview? Item 1 1 is idenchal to item 9 of the preceding questionnaire If you used the information to help you prepare, how did you use it? How did it help you, if at all? LIST OF REFERENCES LIST OF REFERENCES Anderson, L. (1991). Test-retest reliability of the Revised Self-Monitoring Scale over a two-year period. ngchologkal Reports, 68, 1057—1058. Baldwin, M. W., & Holmes, J. G. (1987). Salient private audiences and awareness of the self. Journal of Personalityand Social Paychologv. 52, 1087-1098. Bearden, W. O., Shuptrine, F. K., & Teel, J. E. (1989). Self-monitoring and reactions to image appeals and claims about product quality. Advances in Consumer Researc_h, 1p, 703-711. Berger, C. R. (1988). Planning, affect, and social action generation. In R. L. Donohew, H. Sypher, & E. T. Higgins (Eds), Communication. social cognition. and affect (pp. 93-116). Hillsdale, NJ: Erlbaum. Berger, C. R. (1994). A plan—based approach to strategic communication. In D. E. Hewes (Ed), The cognitive bases of interpersonalfcommunigation. Hillsdale, NJ: Erlbaum. Berger, C. R., & Bell, R. A. (1988). Plans and the initiation of social relationships. Human Communication Research, l_5_, 217-235. Berger, C. R., & DiBattista, P. (1992). Information seeking and plan elaboration: What do you need to know to know what to do? Communication Monoggaphs, 5_9, 368- 387. Berger, C. R., & DiBattista, P. (1993). Communication failure and plan adaptation: If at first you don’t succeed, say it louder and slower. Communication Monogtaphs, 60, 220-238. Berger, C. R., & Jordan, I. M. (1992). Planning sources, planning difficulty, and verbal fluency. Communication Monographs, _5_9_, 130-149. Berger, C. R., Karol, S. H, & Jordan, J. M. (1989). When a lot of knowledge is a dangerous thing: The debilitating effects of plan complexity on verbal fluency. Human Commpnigrgon Research. 16, 91-119. 76 77 Brandt, D. R., Miller, G. R., & Hocking, J. E. (1980). Effects of self-monitoring and familiarity on deception detection. Comrppnication Quarterly, 28, 3-10. Briggs, S. R., Cheek, J. M., & Buss, A. H. (1980). Analysis of the Self-Monitoring Scale. Journal of Personality and Social Psychology, 55, 663-678. Buller, D. B., Comstock, J., Aune, R. K., & Strzyzewski, K. D. (1989). The effect of probing on deceivers and truthtellers. JOULQQI of Nonverbal Belgvior. 1_3, 155-170. Buller, D. B., Strzyzewski, K. D., & Hunsaker, F. G. (1991). Interpersonal deception: II. The inferiority of conversational participants as deception detectors. Commpnfirtion Monogtaphs, M, 25-40. Cody, M. J ., Marston, P. J ., & Foster, M. (1984). Deception: Paralinguistic and verbal leakage. In R. N. Bostrom (Ed), Communicationyearboglfi (pp. 464-490). Beverly Hills, CA: Sage. Cronshaw, S. F., & Ellis, R. J. (1991). A process investigation of self-monitoring and leader emergence. ml] Group Research, _2_2, 403-420. DePaulo, B. M., Davis, T., & Lanier, K. (1980). Planning lies: The effects of spontaneity an_d arousal on success at decgption. Paper presented at the Eastern Psychological Association, Hartford, CN. DePaulo, B. M., Kirkendol, S. E., Tang, J ., & O’Brien, T. P. (1988). The motivational impairment effect in the communication of deception: Replications and extensions. Joupnal of Nonvgrpal Behavior, l_2, 177-201. deTurck, M. A., & Miller, G. R. (1985). Deception and arousal: Isolating the behavioral correlates of deception. Humap Communication Researgh, 1_2_, 181-201. deTurck, M. A., & Miller, G. R. (1990). Training observers to detect deception: Effects of self-monitoring and rehearsal. Human Communication Research, 1_6_, 603-620. Dillard, J. P., & Hunter, J. E. (1989). On the use and interpretation of the Emotional Empathy Scale, the Self-Consciousness Scales, and the Self-Monitoring Scale. Communication Research. _lé, 104-129. Dobbins, G. H., Farh, J. L., & Werbel, J. D. (1993). The influence of self-monitoring on inflation of grade-point averages for research and selection purposes. Journal of Applied Social Psycholpgy, _2_3, 321-334. Duval, S., & Wicklund, R. A. (1972). A theory of objective self-awareness. New York: Academic Press. 78 Duval, S., & Wicklund, R. A. (1973). Effects of objective self-awareness on attribution of causality. Journal of ExperimentaLSociaflsycholpgy, 2, 17-31. Eichenhofer, D., Gerstein, L., Valutis, W., & Jankowski, J. (1987). Effects of anxiety on self-monitoring. Psycholpgical Reports. M, 831-836. Ekman, P. (1988). Lying and nonverbal behavior: Theoretical issues and new findings. Joumal of Nonverbal Behavior. 2, 163-175. Elliot, G. C. (1979). Some effects of deception and level of self-monitoring on planning and reaction to self-presentation. Journal of Personality and Social Psychology, fl, 1282-1292. Friedman, H. S., & Miller-Herringer, T. (1991). Nonverbal display of emotion in public and private: Self-monitoring, personality, and expressive cues. Jouraal of Personality and Social Psychology, 61, 766-775. Greene, J. O. (1984). A cognitive approach to human communication: An action assembly theory. Communication Monoggaphs 51 289-305. ’—9 Green, J. O., & Lindsey, A. E. (1989). Encoding processes in the production of multiple- goal messages. Human Communication Research. 1_6, 120-140. Greene, J. O., O’Hair, H. D., Cody, M. J., & Yen, C. (1985). Planning and the control of behavior during deception. Human Communication Resea_rc_h, _1_1_(3), 335-364. Goldsmith, R. E. (1987). Self-monitoring and innovativeness. Psychological Reports, @, 1017-1018. Hosch, H. M., & Platz, S. J. (1984). Self-monitoring and eyewitness accuracy. Personality and Social Psychology Bulletin. m, 289-292. Hoyle, R. H. & Lennox, R. D. (1991). Latent structure of self-monitoring. Multivariate Behavioral Research, 26, 511-540. Ickes, W., Layden, M. A., & Barnes, R. D. (1978). Objective self-awareness and individuation: An empirical link. Journal of Personality, 59, 146-161. Lassiter, G. D., Stone, J. I., & Weigold, M. F. (1987). Effect of leading questions on the self-monitoring-memory correlation. Personality and Social Psychology Bulletin, M, 537-545. Lennox, R. D., & Wolfe, R. N. (1984). Revision of the self-monitoring scale. Journal of Personalig and Social Psvghology, 56, 1349-1364. 79 Littlepage, G. E., Tang, D. W., & Pineault, M. A. (1986). Nonverbal and content factors in the detection of deception in planned and spontaneous communications. Journal of Social Behavior and Personality, 1, 439-450. Miller, G. R.; deTurck, M. A.', & Kalbfleisch, P. J. (1983). Self-monitoring, rehearsal, and deceptive communication. Human Communication Research, m, 97-117. Mufson, L. & Nowicki, S. (1991). Factors affecting the accuracy of facial affect recognition. Joum of Socialjsychology, 13_1, 815-822. O’Sullivan, M., Ekman, P., & Friesen, W. V. (1988). The effect of comparisons on detecting deceit. Joumal_of Nonverbal Behavior. 2, 203-215. Riggio, R. E., & Friedman, H. S. (1983). Individual differences and cues to deception. Journal of Personality and Socarl Psychology, 4;, 899-915. Siegman, A. W., & Reynolds, M. A. (1983). Self-monitoring and speech in feigned and unfeigned lying. Jouml of Personalig and Social Psychology, M, 1325-1333. Shaffer, D. R., Ogden, J. K., & Wu, C. (1987). Effects of self-monitoring and prospect of future interaction on self-disclosure reciprocity during the acquaintance process. Journal of Personality. 55, 75-96. Shuptrine, F. K., Bearden, W. 0., & Teel, J. E. (1990). An analysis of the dimensionality and reliability of the Lennox and Wolfe Revised Self-Monitoring Scale. Journal of Personalig Assessment, M, 515-522. Snyder, M. (1974). The self-monitoring of expressive behavior. Joarnal of Personalig and Soc_ial Psychology, 3_Q, 526-537. Snyder, M. (1979). Self-monitoring processes. In L. Berkowitz (Ed), Advances in egerimental social psychology (Vol. 12, pp. 85-128). New York: Academic Press. Snyder, M. (1987). Public appearances/private reafities: The psychology of self- monitoring. New York: Freeman. Snyder, M., & Gangestad, S. (1986). On the nature of self-monitoring: Matters of assessment, matters of validity. Journal of Personality and Sociaflsycholog, 51, 1 25-1 39. Snyder, M., & Monson, T. C. (1975). Persons, situations, and the control of social behavior. Journal of Personality anal Sociala Psychology, 3;, 637-644. 80 Stiff, J ., Corman, S., Krizek, B., & Snider, E. (1994). Individual differences and changes in nonverbal behavior: Unmasking the changing faces of deception. Comrraraipation Research, 21, 555-581. Stiff, J. B. & Miller, G. R. (1986). “Come to think of it...”: Interrogative probes, deceptive communication, and deception detection. Human Comunication Research, 12, 339-358. Sypher, B. D., & Sypher, H. E. (1983). Perceptions of communication ability: Self- monitoring in an organizational setting. Personality and Social Psychology Bulletin, 2, 297-304. Turner, R. G. (1980). Self-monitoring and humor production. Journal of Personality, 48, 163-172. Waldron, V. R. (1990). Constrained rationality: Situational influences on information acquisition plans and tactics. Communication Monographs, 5_7_, 184-201. Webb, W. M., Marsh, K. L., Schneiderman, W., & Davis, B. (1989). Interaction between self-monitoring and manipulated states of self-awareness. Journal of Perscaraaity aad Sofl Psychology, 5a, 70-80. Zuckerman, M., DePaulo, B. M., & Rosenthal, R. (1981). Verbal and nonverbal communication of deception. In L. Berkowitz (Ed), Advances in experimental social psychology. Vol. 14 (pp. 2-59). New York: Academic. Zuckerman, M., & Driver, R. (1985). Telling lies: Verbal and nonverbal correlates of deception. In A. W. Siegman and S. Feldstein (Eds), Nonverbal communication: An integrated perspective (pp. 129-147). Hillsdale, NJ: Erlbaum. NOTES NOTES 1 Participants in both Study 1 and Study 2 completed the 18-item Snyder and Gangestad (1986) Self-Monitoring Scale in addition to the Lennox and Wolfe (1984) Revised Self-Monitoring Scale. Although it was possible, with the deletion of three items (items 2, 3, and 11 of the original measure) to produce three internally consistent factors, parallelism could not be achieved for these factors. The first factor (acting) yielded an or of .70. The second factor (other-directedness) produced an a of .71, and the third factor (extroversion/sociability) yielded an or of. 66 Because it proved impossible to obtain a factor structure that did not demonstrate substantial nonparallelism, analyses were not conducted using the Snyder and Gangestad version of the scale. 2 Despite the problems inherent in computing a single scale score for a multidimensional measure that does not demonstrate second order unidimensionality, this procedure was used to maximize comparability with the Miller et al. (1983) procedure. It should be noted, however, that the three dimensions of the self-monitoring scale produced similar correlations with detection accuracy (_r_s ranging from -.11 to -.20). 81 ' "11111111111111111111“